in re: puerto rico electric case no. cepr-ap-2015-0001...

IN RE: PUERTO RICO ELECTRIC POWER AUTHORITY RATE REVIEW

CASE NO. CEPR-AP-2015-0001

SUBJECT: Order requiring revised testimony

ICSE'S MOTION SUBMITTING TESTIMONIES

TO THE ENERGY COMMISSION:

NOW COMES the Institute de Competitividad y Sostenibilidad Econ6mica de Puerto Rico (ICSE-PR) represented by appearing counsel and respectfully alleges and prays:

1. Together with this Motion ICSE-PR is submitting four testimonies:

a. Ramon J. Cao Garcia- Economist President of Asesoria y Consultoria, Inc. b. Victor Glass -Director of Rutgers Center for Research in Regulated

Industries (CRRI) c. Tom Sanzillo-Director of Finance for the Institute for Energy Economics

and Financial Analysis and Cathy Kunkel- Energy Analyst d. Fernando E. Agrait-Attorney

2. The sworn testimonies are in compliance with PREC instructions to intervenors.

WHEREFORE, ICSEPR respectfully requests that the testimonies be received and accepted.

I certify that I have served a PDF copy of the foregoing to the following persons: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; m mu ntanerlaw@gmail. com; jfelicia no@constructorespr. net; abogados@fuerteslaw. com; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; jorgehernandez@escopr. net; ecandelaria@camarapr. net; pga@caribe. net; manuelgabrielfernandez@gmail. com; mreyes@m idapr.com; mgrpcorp@gmail. com; [email protected]; [email protected] and [email protected].

1

/ /I rz I f J ;1

In San Juan, Puerto Rico, at 25 days of~t:to)jer 2016. //

//// !I 1 111 •. ,;:l

c./F;( 1\.. ,j .

1 .. t

1t/}J\ .... '

I !\ / '\ I ... ! I

(//\.. \ l•'•i/·

, ! '<, J I v /FERNANDO "E::/AGRAI

I ' )".S. NUM. 3772

/V01 AVENIDA PONCE DE LEON /)OFICINA414

{/SAN JUAN, PUERTO RICO 00907 TELS 787·725-3390/3391 FAX 787-724-0353 EMAIL: [email protected]

2

1 TESTIMONY OF RAMON J. CAO GARCIA

2 I. INTRODUCTION

3 A. Witness Identification

ReciWM .-w.-~..;..,;,_;...~-.a.::~~--

feeha: -~-:;.;;:;;J...;......J..~;;...---

4 Q. Please state your name, title, employer, and business address.

5 A. I am Ramon Cao Garda, President of Asesorfa y Consulta, Inc., located at Urb.

6 Cupey Gardens, C-11 Calle 3, San Juan, PR 00926.

7 Q. On whose behalf are you testifying?

8 A. I am testifying on behalf of the Instituto de Competitividad y Sostenibilidad

g· Econ6mica de Puerto Rico (ICSEPR).

10 B. Summary of Testimony

11 Q. What is the purpose of your testimony?

12 It purpose is to assess expected economic consequences of proposed

13 electricity tariff rates increases. The economic topics considered are: (1)

14 consequences of proposed rates on production costs in eight economic sectors,

15 (2) their effect on Puerto Rico's Real GNP and total employment, (3) the impact

16 of proposed new rates on cost ofliving, and (4) expected consequences on the

17 quantity of electricity demanded.

18 C. Professional Background & Education

19 Q. Would you please describe your educational background and professional

20 experience?

Page 1 of13

21 A

22

23

24

25 Q.

26

27 A

28

29 II. 30

31 Q.

32

33 A

34

35

36

37

38

39

40

41

42

Yes. My curriculum vitae, which reviews my education, professional

qualifications, publications, honors received, and experience in detail, is

attached as Exhibit 1.

D. Additional Attachments

In addition to your curriculum vitae, are there any additional exhibits to your

testimony?

Yes. This testimony includes the following attachments:

Exhibits 1, 2, 3, 4, 5, 6, 7, 8, and 9

EFFECTS OF PROPOSED RATE INCREASE ON THE PRODUCTIVE SECTORS OF THE ECONOMY

Why are you testifying on economic consequences of PREP A's proposed tariff

rates increases?

I examined many documents produced by PREP A in support of the proposed

increase in electricity tariff rates. It surprised me that important and relevant

information was not included in the documents examined. For example, no

information is provided, nor consideration appears to be given, to the

Historical Marginal Energy Costs, and a subjective and arbitrary 5%

differential threshold is established for all rates except public lighting, with no

objective criteria. Furthermore, in the documents examined, I was not able to

find any consideration about economic consequences of rate increases on the

local economy, or even on how they are expected to affect the consumption of

electricity. It is of foremost importance to, at least, have an idea about the

Page 2 of13

43

44

45

46

47

Q.

48 A.

49

50

51

52

53 Q.

54 A.

55

56

57

58

59

60

61

consequences of PREP A's proposal on the productive sectors of the economy,

as well as in cost of living of residents; since electricity is an essential

production input, and a necessity for everyday contemporary civilized life.

What sort of effects could have PREP A's proposed tariff rates increases on the

productive sectors of the economy?

Since electricity is an intermediate input for all productive sectors in the

economy, an increase in its price will raise production costs. This will reduce

the ability of producers to compete in local and export markets, with

consequences upon the total output in the economy, and, hence, on

employment levels in the local labor market.

How these effects were measured?

To estimate expected consequences of PREP A's proposed increases in tariff

rates on the industrial sectors of the economy, it was obtained, from the PR

Planning Board, the 2013 Input/Output Transactions Matrix. This is an update,

using the RAS method, of the 2002 matrix. The matrix was condensed, from its

original 93 industrial sectors, to 8 industries, which arel:

1.

2.

3.

Agriculture

Mining and Construction

Manufacturing

1 See Exhibit 2.

Page 3 of13

62

63

64

65

66

67

68

69

70

71

72

73

Q.

A.

4. Commerce

5. Hospitals

6. Electric Production

7. Other Services

8. Government

The electric production vector in the condensed transactions matrix was

modified, increasing it by PREPA's proposed increase in rates for each

sector.2,3 It was possible to compute the expected change in costs by industrial

sector, as well as it overall impact in costs.

Which were the results from these estimations?

The increase in intermediate inputs costs, among the eight economic sectors

considered, are as follows:

Economic Sector

Agriculture

Mining & Construction

Manufacture

Wholesale & Retail Trade

Increase in Cost of

Intermediate Inputs 0.12%

0.25%

0.51 o/o

1.33%

z See Exhibits 3 and 4.

3 It shonld be noted that considered increases, of proposed electric energy costs, in the Input/Output Matrix do not include the proposed increase in residential tariff rates.

Page 4 of13

74

75

76

77

78 Q.

79

80 A.

81

82

83

84

85

86

87

88

Hospitals & Health Serv. 0.25%

Electricity 0.00%

Other Services 0.38%

Government 1.09%

Overall 0.53%

The most affected sectors are: (1) wholesale and retail trade, with an increase

in cost of intermediate inputs of 1.33%; (2) government, with a 1.09%

increase; and (3) manufacture, which face an estimated increase of 0.51 %.

Which are the economic consequences of expected increases in the cost of

intermediate inputs?

Economic sectors producing for the local market may pass along the increase

in costs by raising prices, if market conditions allows. Export industries or

those where local market conditions do not allow for price adjustments, will

have to absorb the increase in costs, reduce output, or close operations in the

case of marginal firms. In consequence, it is reasonable to expect that PREP A's

proposed increase in tariff rates will tend to generate an increase in the

Consumer Price Index, increasing cost of living and reducing real income or

purchasing power among the population; and a further contraction in real

GNP.

Page 5 of13

89 Q.

90

91 A.

92

93

94

95

96

97

98

99

100 III. 101

102

103 Q.

104

105 A.

106

107 Q.

108

Which economic sectors will be less able to adjust price to face the increase in

intermediate inputs cost?

Industrial sectors facing the most adverse conditions to be able to adjust

prices are:

~ Agriculture, which faces stiff competition from imports.

~ Mining and construction, because of present adverse conditions in the

real estate market.

~ Manufacture, due to its high concentration in export production.

$ Hospital and health services, because it mostly provide services to

insured patients, and have to accept the service fees established by

insurance companies.

IMPACT OF PROPOSED ELECTRICITY TARIFF RATES ON THE LEVEL OF ECONOMIC ACTIVITY

A. Impact on GNP

Did you evaluate the impact of proposed higher electricity tariff rates on the

level of economic activity?

Yes. I estimated the expected effects of the proposed tariff rates on real GNP,

and on the employment level.

How do you estimated expected effects of the proposed tariff rates on real

GNP?

Page 6 of13

109 A

110

111

112

113 Q.

114 A

115

116

117

118

119

120

121

Q.

122 A

123

124

125

126

127

To simulate the impact of proposed increases in electricity rates on real GNP,

a forecasting econometric equation was constructed and estimated through

the Cochrane-Orcutt method. Exhibit 5 informs the estimation equation, data

used, and results obtained.

Which is the effect that you estimated, on real GNP?

Using the regression coefficient for the price of electricity, and assuming an

increase of 4.2¢ per Kwh in the price of electricity (as per Schedule H), it is

estimated reduction of $66.2 million en real GNP,4 which is equivalent to a

reduction of 1.05% in real GNP. This decline is additional to the 2.0% decline

forecasted by the PR Planning Board for real GNP in fiscal year 2016-2017.

B. Employment Effects

You said that you also estimated expected effects of proposed electricity rates

on employment levels in Puerto Rico. How you did such estimation?

First, it was computed the employment requirements per million dollars of

real GNP. The values in the series were averaged, and the resulting average

was multiplied by the estimated reduction of $66.2 million in real GNP. The

result is the expected effect on employment of the proposed increase in the

price of electricity. Exhibit 6 informs the data used and the computations

performed.

4 Real GNP in Puerto Rico is estimated in dollars of 1954.

Page 7 of13

128 Q.

129 A.

130

131

132

133

134

135

136

137 IV. 138

139 Q.

140

141

142 A.

143

144 Q.

145 A.

146

What is the employment effect you estimated?

The average relation between Employment and real GNP in Puerto Rico, for

fiscal years 2000 to 2016, is 167 jobs per million dollars of real GNP. Ifthere is

$66.2 million contraction in real GNP caused by the proposed increase in

electricity rates, then it is reasonable to expect a reduction of 11,075 jobs. To

illustrate its significance, actual unemployment rate in August, 2016,

according to the Puerto Rico Labor and Human Resources Department,s was

12.1 %; an increase in unemployment of the magnitude of 11,075, would had

raised the unemployment rate to 13.1 %, a whole percentage point increase.

EXPECTED CONSEQUENCES UPON THE COST OF LIVING OF THE POPULATION

Did you considered in your economic analysis the expected consequences of

the proposed rate increase in electricity rates upon the cost of living of the

general population?

Yes, I estimated the effects of the proposed rate increases in the price of

electricity on the Consumer Price Index.

How you made such an estimate?

Computation of the effects of the proposed rate increases in the price of

electricity on the Consumer Price Index was made as follows:

s Puerto Rico Labor and Human Resources Department, Empleo y Desempleo en Puerto Rico, August 2016, p. 10, Table 1A.

Page 8 of13

147 1. For electricity cost the residential rate increase was multiplied by the price

148 index weight.6 It was conservatively assumed that there are going to be no

149 other increases in prices in the Housing and Housing Services Sector.

150 2. For the categories of apparel, food and beverages and other goods and

151 services, the cost increase computed using the Input/Output Matrix for

152 Wholesale and Retail Trade were multiplied by the price index weight.

153 3. For education and communications, and entertainment, the price increase

154 was computed using the increase in costs estimated from the Input/Output

155 Matrix for Other Services and multiplied by the price index weight.

156 4. For Health Services the cost increase computed through the Input/Output

157 Matrix for Hospitals and Health Services was multiplied by the price index

158 weight.

159 5. In the case of the Transportation Sector it was assumed that there are not

160 going to be price increases; recognizing that it is a conservative

161 assumption.

162 Exhibit 7 presents the data used and computations performed.

163 Q. Which were your results?

164 A. The resulting estimate is an expected increase of 1.35% in the Consumer Price

165 Index.

6 The price index weights used in all computations were those computed by the Puerto Rico Labor and Human Resources Department and the Statistics Institute of Puerto Rico for December 2006.

Page 9 of13

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

Q.

A.

v.

Q.

183 A.

Could you summarize the effects you estimate for the proposed increase in

prices of electricity on the general population?

It is clear that the proposed increase in electricity tariff rates is expected to

have significant consequences on the general population:

Real GNP is expected to decline by 1.05%, in addition to the 2.0%

contraction forecasted by the PR Planning Board for fiscal year 2017. This

implies a significant reduction in the income available to residents.

Cost of living is expected to increase by 1.32%, which implies a

corresponding decline in the purchasing power of the population.

In consequence, PREPA's proposed rate increase will result in making the

general population 2.36% poorer, aggravating conditions created by the

general contraction faced by the Puerto Rican economy since fiscal year 2007,

including the 2.0% contraction in real GNP forecasted by the Puerto Rico

Planning Board for fiscal year 2017.

EXPECTED CONSEQUENCES ON THE DEMAND FOR ELECTRICITY

Did you considered any additional general economic consequence of the

proposed electricity tariff rate increase?

Yes. I examined some of the testimonies and exhibits submitted by PREP A in

184 support of its proposal to increase tariff rates, and they seem to assume that

185 the quantity of electricity demanded will stay at the level registered in fiscal

186 year 2014 (Refer to Schedule F in this proceeding). Such assumption is

Page 10 of13

187 unrealistic, and clearly adversely affects all estimates PREP A made about the

188 forecasts of additional revenues allegedly required by PREP A. It appears that

189 PREP A did not took into account the trend shown in the quantity of electricity

190 demanded since fiscal year 2007, nor the expected effects of a change in price

191 on the demand for electricity.

192 Q.

193 A.

194

195

196

197

198

199

200

201

202

203

204 Q.

205

206 A.

207

Which is the trend in the quantity of electricity demanded?

As it is shown in Exhibit 8 to this testimony, PREPA's net production of

electricity declined by 3,322.6 million Kwh, or 16.7%, between fiscal years

2007 and 2016. The steep decline in the demand for electricity that is

happening in Puerto Rico since present economic contraction began in fiscal

year 2007 it is something that should be taken into account in the design of

any tariff revision. It is simply unrealistic to assume that the demand of

electricity will somehow remain at the level registered in fiscal year 2014,

particularly when, accordingly to PREP A's data, net production declined by

211.8 million Kwh, or 1.21 %, between fiscal years 2014 and 2016. It should be

also noted that an increase in price, as proposed, will have an effect reducing

quantity demanded.

How it can be forecasted the effect of an increase in tariff rates on the quantity

of electricity demanded ?

To forecast this effect it is necessary to estimate a demand equation for

electricity. Ideally, demand equations by consumer categories should be

Page 11 of13

208

209

210 Q.

211 A.

212

213

214 Q.

215 A.

estimated, because, conceptually, different types of consumers can have

different price and income elasticities.

Did you estimate any demand equation?

Yes, I estimated an aggregate demand equation for elasticity. Time and

resources constraints prevented the estimation of demand equations by

consumer categories.

How did you estimate that demand equation?

To estimate the effect of proposed increases in electricity rates on the quantity

216 of electricity demanded, a demand equation for electricity was constructed

217 and estimated through the Two Stages Least Squares method (ZSLS). Exhibit

218 9 informs the estimation equation, data used, and results obtained.

219 Q.

220

221 A.

222

223

Which is the effect you estimated that will have an increase in electricity prices

on the quantity of electricity demanded?

Assuming an increase of $0.042 in the price of Kwh and a population of 3.4

million, it is estimated that quantity of electricity demanded will decline by

144.4 million Kwh, or 0.83%.

224 VII. CONCLUSION

225 Q. Does this complete your direct testimony?

226 A. Yes.

Page 12 of13

227

228

229

230

231

232

233

234

',/) ; I Y~-~.~~

RAMON J. CAO GARCIA

Affidavit No. :;v)'~

Sworn before me by Ramon J. Cao Garcia, economist, of legal age, married

and resident of San Juan, Puerto Rico to whom I known personally, today

~ :2"/"bf October of 2016.

0 <:J

OS ~

0

i 1·: .. :'· .. -~ ... ~' --'

Sello ·.

In San Juan, Puerto R .. ico. }·.

; I

\ y ./

Page 13 of13

Exhibit1

POSTAL ADDRESS

LANGUAGE ABILITY

ACADEMIC DEGREES

FIELDS OF INTEREST

HONORS

CURRICULUM VITAE

Ramon J. Cao Garcia Urb. Cupey Gardens, C-11, 3rd St.

Rio Piedras, Puerto Rico 00926-7306

Urb. Cupey Gardens C-11 Calle 3 San Juan, PR 00926-7306

Spanish English Italian Portuguese

Ph.D. - Economics

(excellent) (excellent) (reading ability) (reading ability)

Virginia Polytechnic Institute and State University (January 1979)

M.A. - Economics University of Puerto Rico, Rio Piedras Campus (June 1971)

B.A. - Economics University of Puerto Rico, Rio Piedras Campus (June 1968)

Statistics and Econometrics Public Sector Economics Urban Economics Human Resources/Economics of Education Forensic Economics

Honor Student Rio Piedras Campus, University of Puerto Rico (1964-1971)

B.A. Magna Cum Laude Rio Piedras Campus, University of Puerto Rico

Page I of 15

HONORS (Cont.)::

(June 1968)

M.A. Magna Cum Laude Rio Piedras Campus, University of Puerto Rico (June 1971)

Scaife fellow in economics Virginia Polytechnic Institute and SU (1971-1972)

Beta Gamma Sigma (National Scholastic Society for Students of Business and Management)

Invited to become a member of the International Association of Public Finance (September 1979)

Member of the Governor's Committee on Economic Policy for Development and Chairman of the Education and Economic Growth Subcommittee (1978-1980)

Member of the Board of Directors of Puerto Rican Economic Association (1984-1985 and 1994-2000)

Member, Private Industry Council on Strategic Planning (1988-1990)

Member of the Editorial Board of Revista de Ciencias Sociales.

Member of the May 8. 2006 Commission (to solve the closing of the Executive Branch of the Government)

Member of the Council of Economic Advisors to the Governor (2009-2012)

Referee for the following professional journals: Caribbean Studies Revista de Administraci6n Publica Revista/Review lnteramericana Revista de Administraci6n de Empresas Revista de Ciencias Sociales

Member Academia Cientffica y Literaria lberoamericana (since 2004)

Page 2 of 15

Ramon J. Cao Garcia Curriculum Vitae

PROFESSIONAL

Member Academia de Artes y Ciencias de Puerto Rico (since 2007)

2014 Annual Assembly of the P.R. Economists Association was dedicated to honor my lifework.

ORGANIZATIONS: -International Institute of Public Finance (not active at present time) -Western Economic Association (not active at present tirne)

TEACHING EXPERIENCE:

ADMINISTRATIVE EXPERIENCE:

-Public Choice Society (not active at present time) -Intercollegiate Studies Association (not active at present time) -Caribbean Studies Association (not active at present time) -Asociaci6n de Economistas de Puerto Rico

Department of Economics Rfo Piedras Campus University of Puerto Rico (since July 1973- Retired June 2004)

Professor Associate Professor Assistant Professor Instructor

Instructor Department of Economics College of Social Sciences Rfo Piedras Campus University of Puerto Rico (Summer 1972)

since 1988 1981-1988 1978-1981 1973-1978

Lecturer in Economics and Statistics Universidad Central de Bayam6n Bayam6n, Puerto Rico (1969-1971 and Summer 1972)

Assistant Dean for Academic Affairs College of Social Sciences Rfo Piedras Campus University of Puerto Rico (January 1983 to July 1985)

Page 3 of 15

Ramon J. Gao Garcia Curriculum Vitae

RESEARCH EXPERIENCE:

Director of the Economic Research Unit, College of Social Sciences Rfo Piedras Campus University of Puerto Rico (August 1982 to July 1983)

Chairman of the Economics Department, College of Social Sciences Rfo Piedras Campus University of Puerto Rico (July 1980 to August 1982)

Advisor to various private enterprises and associations in Puerto Rico.

Consultant to various government agencies in Puerto Rico, including, among others, the following: ·

Puerto Rico Tax Reform Commission (1973-1975)

Puerto Rico Treasury Department (1975-1976)

Puerto Rico Planning Board (1978)

Puerto Rico Right to Work Administration (1979)

Commission on the Social Responsibility of the University of Puerto Rico

(1984-1985)

President of the Senate of Puerto Rico (1986-1989)

Puerto Rico Manufacturer's Association (1979-1990)

Puerto Rico Chamber of Commerce Page 4 of 15


RESEARCH EXPERIENCE (Cont.)

(1990-2005)

President of the University of Puerto Rico (1990-1993)

Secretary of the Treasury- Tax Reform (1993-1995)

Secretary of the Treasury of Puerto Rico (1995-2001)

Commerce Development Administration (1998-2000)

Office of the Comptroller of Puerto Rico (200 1-2005, 2006-2010)

Department of the Treasury of Puerto Rico (fiscal reform) . (2005-2006)

Consultant to the United Nations Development Program on the following projects:

-Economic Development and Planning for El Salvador Planning Council (1975)

-Methodology for a Plan for the Development of the Niger River Basin (1976-1977)

-Modernization of the Venezuelan Fiscal System (1977)

-Cost/Benefit Analysis of the Cartagena Commercial and Industrial Free Trade Zone (1977-1978)

Economist of the Caribbean Economic Development Corporation (1969)

Research Assistant Economic Researh Unit College of Social Sciences Rfo Piedras Campus

Page 5 of 15


University of Puerto Rico (1969-1971)

Research Assistant (part-time) Public Choice Center Virginia Polytechnic Ins!. & S.U.(1972-1973)

CONSULTING EXPERIENCE IN THE PRIVATE SECTOR (SELECTED PROJECTS): Co-editor of EGO-NEWS and consultant to the Puerto Rico

Manufacturers Association on various projects. EGO-NEWS was a quarterly newsletter written for Puerto Rican Manufacturers Association. It analyzed economic conditions in Puerto Rico.

(1979-1990) Subcontracted by Clapp & Mayne, Inc. to work on various projects, including a feasability study for the Chase Manhattan Bank, and various financial and managerial rehabilitation studies of clothing factories for the Trade Adjustment Assistance Center, monitored by the Institute of Public Administration.

Consultant to various private enterprises, including:

Crown Cork of Puerto Rico (market research) (1979)

Mafresu, Inc. (financial feasability) (1980)

Omark Caribbean (statistical quality control) (1981)

General Electric of Puerto Rico (trends of prices and costs in the Puerto Rico economy) (1983)

California Rice Growers Association (market and demand analysis and forecasts for 14

Page 6 of 15


PUBLICATIONS: (Selected List)

commodity categories) (1987)

Cervecerfa India (demand for local and imported beer in Puerto Rico) (1989)

Books

Contribuci6n sabre los ingresos de las personas en Puerto Rico (with S. Andie and F. Andie). San Juan, Puerto Rico: Comisi6n de Reforma Contributiva de Puerto Rico, 197 4 (out of print).

Principios y metodos estadfsticos para comercio y economfa (with J. R. Stockton and C. T. Clark). Cincinnati, Ohio: South-Western Publishing Co., Volume I, 1980; Volume II, 1982.

Economfa basica (Spanish translation ofT. J. Hailstones, Basic Economics). Cincinnati, Ohio: South-Western Publishing Co., 1982.

Explorations Toward an Economic Theory of Political Systems. Lanham, Md.: University Press of America, 1983.

Economfa de Ia reforma fiscal en Puerto Rico (Ramon J. Cao Garda, ed.) Cincinnati, Ohio: South-Western Publishing Co., 1983.

Metodos estadfsticos: teorfa y practica (with J. R. Stockton and C. T. Clark). Chicago: Scott and Foresman Publishing Co., 1985.

Ensayos econ6micos (edited with A L. Ruiz y F. Zalacafn). Proceedings of the Puerto Rican Economic-Association Annual Conference of 1984. San Juan, 1986.

Educaci6n universitaria y oportunidad econ6mica en Puerto Rico (with H. Matos). Madrid: Editorial Betania, 1988.

Page 7 of 15


Reform a Contributiva en Puerto Rico 1994 - Estudio Tecnico (edited with S. Andie). San Juan: Editorial de Ia Universidad de Puerto Rico, 1996.

Estrategias para el desarrollo: Competitividad, productividad y eficiencia (editor). San Juan: Camara de Comercio de Puerto Rico, 1997.

lmpuestos en Puerto Rico: Treinta afios de experiencias y estudios. San Juan: Grupo Editorial Akron/Sistema Universitario Ana G. Mendez, 2004.

Professional Articles

1970

"Cooperativismo y desarrollo". Nueva Generaci6n, Vol. Ill, No.2, November 1970.

1979

"Accidentes de trans ito: un modelo econ6mico" (with H. Matos). Revista de Ciencias Sociales, Vol. XXI, No. 1-2, March-June 1979.

"The Environment and the Guidelines for Project Evaluation" (with F. Andie), in W. Beller, ed., Proceedings of the Conference on Environmental Management and Growth in the Smaller Caribbean Islands. Wilday, St. Michael, Barbados. Department of State Publication 8998, U.S. Government Printing Office, September 1979. (Reprinted by the Economic Research Unit, Rfo Piedras Campus, University of Puerto Rico).

"La teorfa econ6mica de los sistemas politicos: evidencia empfrica". Revista de Administraci6n Publica, Vol. XII, No. 1, October 1979.

"Distribuci6n del ingreso en Puerto Rico: unos comentarios y un nuevo analisis". Revista de Ciencias Sociales, Vol. XXI, No. 3-4, SeptemberDecember 1979.

1980 Page 8 of 15


"Cost and Benefit of a Free Zone: The Case of Cartagena" (with F. Andie). Caribbean Studies, Vol. XX, No. 1, March 1980. (Reprinted by the Economic Research Unit, Rio Piedras Campus, University of Puerto Rico).

"Az Gelismis Ulkederle Vergi Reformlarinim Trend Ve Fonksiyonlari: Bazi Sinirlayici Facktorler" (with F. Andie), in Nihad S. Sayar'a Armagan, et.al., eds., Gagdas Vergilermediki Son Calismeler. lstambul: 1980.

1981

"Educaci6n privada y desigualdad: un analisis econ6mico". Revista de Ciencias Sociales, Vol. XXIII, No. 1-2, March-June 1981.

"Trends and Functions of Tax Reforms in LDC's: Some Limiting Factors" (with F. Andie), in K. W. Roskamp and F. Forte, eds. Reforms of Tax Systems. Detroit: Wayne State University Press, 1981. (Reprinted by the Economic Research Unit, Rio Piedras Campus, University of Puerto Rico).

"Migraci6n interna en Puerto Rico: un analisis econ6mico". Revista de Administraci6n Publica, Vol. XIV, No. 1, October 1981.

1982

"Puerto Rico: recesi6n o crisis". Analisis- Revista de Planificaci6n. Vol. 1, No. 1, January-June 1982.

"Condiciones de vida en los municipios de Puerto Rico: un analisis preliminar". Schemas, Vol. I, No. 1, October 1982.

"EI poder politico de Ia burocracia en Puerto Rico: un analisis preliminar". Revista de Administraci6n Publica, Vol. XV, No. 1, October 1982.

"Economia de Puerto Rico: una interpretacion de sus problemas para los afios 80". Revista Juridica de Ia Universidad de Puerto Rico. Vol. Ll, No.3-4, 1982.

1983

"Reforma fiscal: concordancias y divergencias", in R. Cao, ed., Economia de Ia reforma fiscal en Puerto Rico. Cincinnati: South-Western Publishing

Page 9 of IS


Co., 1983, pp. 2-14.

"lmpuestos de base amplia y reforma fiscal", in R. Gao, ed., Economfa de Ia reforma fiscal en Puerto Rico, pp. 54-67.

"Un analisis economico de Ia educacion publica en Puerto Rico" (with A. Ortiz). Schemas, Vol. I, No.3, January 1983.

1984

"Regulation in a Democracy: Inefficiency or Efficiency" (with S. Andie), in H. Hanusch, ed., Public Finance and the Quest for Efficiency. Detroit: Wayne State University Press, 1984.

1985

"lmpacto economico de Ia Universidad de Puerto Rico: afio fiscal 1982-1983" (with A. Rivera y A. L. Ruiz). lnforme de Ia Comision de Responsabilidad Social de Ia Universidad de Puerto Rico, Vol. II. Rfo Piedras: Universidad de Puerto Rico, 1985. (Reprinted by the Economic Research Unit, Rfo Piedras Campus, University of Puerto Rico).

"Un analisis de las condiciones salariales del personal docente de Ia Universidad de Puerto Rico". Revista deAdministracion Publica, Vol. XVIII, No. 1, October 1985.

1986

"Un analisis del consume en Puerto Rico", en R. Gao, A. L. Ruiz y F. Zalacafn, eds., Ensayos Economicos. San Juan: Asociacion de Economistas de Puerto Rico, 1986, pp. 39-50.

"Informacion acerca de empresas pequefias y medianas en Puerto Rico" (with A. Mann). Comercio y Produccion, Vol. XXVI, No.4, July-August 1986.

1987

"La decision de tener hijos: un modele economico de mujeres casadas." Revista/Review lnteramericana, Vol. XVII, No. 1-2, Spring-Summer 1987.

"Demanda por recursos humanos y planificacion educativa". Plerus, Vol. XX, No. 1-2, June-December 1987.

Page 10 of 15


1988

"University Education as an Homogeneizing Process: An Exploratory Analysis" (with H. Matos), Journal of Economics of Education, 1988.

1989

"Tasa de rendimiento privado de Ia educaci6n: Un estimado para Puerto Rico" (with H. Matos), Revista/Review lnteramericana, Vol. XIX, No. 1-2, Spring/Summer 1989.

1991

"Educaci6n universitaria y realidad social: mirando al futuro". Realidad social puertorriquena y reconceptualizaci6n del currfculo universitario. San Juan: Sistema Universitario Ana G. Mendez, 1991.

"Modelos de planificaci6n educativa", Plerus

1992

"Reforma municipal y finanzas de los municipios". Serie de Publicaciones, Unidad de Investigaciones Econ6micas, Universidad de Puerto Rico, Recinto de Rio Piedras, May 1992.

"EI Acuerdo de Libre Comercio de Norteamerica y Ia competitividad de las empresas puertorriquenas", Comercio y Producci6n, Sept.-Oct. 1992.

1995

"Transportaci6n y progreso", Comercio y Producci6n, Sept.-Oct. 1995.

"Perspectivas empresariales: Resultados de una encuesta a los socios de Ia Camara de Comercio de Puerto Rico", Boletfn AEPR, Vol. IX, No.2, pp. 5-6.

"Cam bios a Ia Secci6n 936: Opiniones de empresarios", Boletfn AEPR, Vol. IX, No.2, pp. 7-9.

Page ll oflS


2000

"Buscando una definicion de Ia 'Nueva Economfa"' Boletfn de Economfa, Unidad de Investigaciones Economicas, UPR, Recinto de Rio Piedras (2000).

2009

"Tariffs and economic consequences of the Government of Puerto Rico experience from the purchasing power agreements for electric generation and energy capacity from privately-owned qualified cogeneration facilities" (with Jose J. Cao Alvira), Proceedings of the B1h Latin-American Congress on Electricity Generation and Transmission- CLAGREE 2009. ISBN 978-85-61065-01-0.

2010 Estudio de Reforma Contributiva Integral en Puerto Rico, (with Juan Lara Fontanez and Jose J. Gao Alvira), San Juan, P.R.: Fundacion del Colegio de Contadores Publicos Autorizados, September 21, 2010, 403 pages.

"Incentives contributivos e inversion privada en Puerto Rico", Revista Civilizar de Empresa y Economfa, Universidad Sergio Arboleda, Bogota, D.C., Colombia, Num. 2, Afio 2, julio-diciembre de 2010, pags. 54 a 74.

2011

"Empirical estimation of avoided costs of coal-fired electric energy cogeneration." (with Jose J. Cao Alvira), Proceedings of the 81h LatinAmerican Congress on Electricity Generation and Transmission- CLAGREE 2011.

Book Reviews

"William Niskanen, Jr.; Bureaucracy and Representative Government". Revista deAdministracion Publica, Vol. VI, No.2, March 1974.

"James M. Buchanan and Richard Wagner; Democracy in Deficit: The Political Legacy of Lord Keynes". Revista de Ciencias Sociales, Vol. XX, No. 2, September 1978.

Page 12 of 15


"Thomas E. Borcherding, ed.; Budgets and Bureaucrats: The Sources of Government Growth". Revista de Administraci6n Publica, Vol. XI, No. 1, October 1978.

"Fuat M. Andie and S. B. Hendrickson, eds.; Readings in Caribbean Public Sector Economics". Caribbean Studies, Vol. XXI, No.2, Apri11981.

"Gordon Tullock, ed.; Toward a Science of Politics: Papers in Honor of Duncan Black". Revista de Ciencias Sociales, Vol. XXIII, No. 1-2, MarchJune 1981.

"International Monetary Fund; World Economic Outlook". Revista/Review lnteramericana, Vol. XVIII, No. 1-2, Spring/Summer 1988.

"Michael Hammer and James Champy, Reengineering the Corporation: A manifesto for business revolution", Boletfn AEPR, Vol. IX, No.1, pp. 12-13.

Working Papers and Monographs

"Equilibria y analisis econ6mico". Economic Research Unit, University of Puerto Rico, Rio Piedras, 1977.

"Zona franca industrial y comercial de Cartagena: un analisis de costa y beneficia". Economic Research Unit, University of Puerto Rico, Rio Piedras, 1978".

"Las fluctuaciones econ6micas y sus efectos sabre el desarrollo de Ia economia". Consejo Estatal de Empleo y Adiestramiento, 1980.

"Metodologia y posibilidades de Ia economia". Economic Research Unit, University of Puerto Rico, Rio Piedras, 1980.

"Elementos en Ia planificaci6n de programas de empleo y adiestramiento". Consejo Estatal de Empleo y Adiestramiento, 1981. "lnestabilidad econ6mica: causas y efectos". Consejo Estatal de Empleo y Adiestramiento, 1981.

"Pasos a seguir para identificar Ia demanda por programas laborales: las necesidades de Ia poblaci6n". Consejo Estatal de Empleo y Adiestramiento, 1981.

Page 13 of 15

Ramon J Gao Garcia Curriculum Vitae

"Teoria de Ia localizaci6n y mercados laborales". Consejo Estatal de Empleo y Adiestramiento, 1981.

"Doctrinas, revoluciones, contrarrevoluciones y el enfoque de oferta", in S. Torres, ed., Analisis econ6mico del enfoque de oferta. Economic Research Unit, University of Puerto Rico, Rio Piedras, 1981.

"Las nuevas !eyes contributivas de 1985", in R. Gao, ed., 1,Reforma contributiva en Puerto Rico?. Economic Research Unit, University of Puerto Rico, Rio Piedras, 1985.

Una evaluaci6n del sistema contributivo de Puerto Rico (with S. Andie) Secretaria del Senado de Puerto Rico, San Juan, 1986.

An Estimate of the Impact of the Minimum Wage on the Economy of Puerto Rico. with Special Reference to Expected Impacts of a Possible Increase in the Minimum Wage and the Manufacturing Sector (with S. Andie). Puerto Rico Manufacturers Association, San Juan, 1987.

"lmpacto econ6mico del Recinto de Rio Piedras de Ia Universidad de Puerto Rico: Ano fiscal1986-87". Oficina de Planificaci6n y Desarrollo, Recinto de Rio Piedras, Universidad de Puerto Rico, 1987.

Anal isis de las tarifas de matricula en Ia Universidad de Puerto Rico. Oficina de Planificaci6n y Desarrollo, Administraci6n Central, Universidad de Puerto Rico, 1991.

"lmpacto econ6mico del Programa de Mejoras Permanentes de Ia Universidad de Puerto Rico". Oficina de Planificaci6n y Desarrollo, Administraci6n Central, Universidad de Puerto Rico, 1991.

"Proyecciones de ingresos al fonda general de Ia Universidad de Puerto Rico: 1992-93 a 1995-96". Oficina de Planificaci6n y Desarrollo, Administraci6n Central, Universidad de Puerto Rico, 1991.

Los Estudiantes de Sociales: Ana/isis de una encuesta a estudiantes de Bachillerato en Ciencias Sociales. Universidad de Puerto Rico, Recinto de Rio Piedras, Facultad de Ciencias Sociales, noviembre de 2001, 152 + iv pages.

PERSONAL DATA: Born on La Habana, Cuba, at October 26, 1947 and living in Page 14 of 15


Revised: March 2015

Puerto Rico since March 1961. Married with three sons.

Page 15 of 15

Exhibit2

Eight Sectors Transaction Matrix Puerto Rico: FY 2013

($thousands) Electricity

Hospitals & Mining& Wholesale& & Health Irrigation Other Total

Agriculture Construction Manufacture Retail Trade Serv. Serv. Services Government Intermediate

Agriculture 357,805 41,461 918,559 51,857 4,200 61 203,026 40,624 1,617,593

Mining & Construction 7,196 80,702 716,221 307,442 64,383 1,433,170 73,292 2,682,406

Manufacturing 340,202 1,591,546 19,722,328 801,313 1,137,327 312,320 4,520,069 1,606,847 30,031,952

Wholesale & Retail Trade 153,090 194,928 5,201,743 268,781 1,053 1,362,585 306,350 7,488,530

Hospitals & Health Serv. 4 1 8 384,304 605 9,052 337,663 731,637

Electricity & Irrigation Serv. 4,935 34,253 563,710 369,360 33,155 218,094 490,493 144,920 1,858,920

Other Services 26,173 1,014,027 1,554,122 4,539,432 1,026,312 51,487 20,054,734 976,664 29,242,951

Government I 589 74,772 109,184 55,006 4,562 374,472 80,688 699,273

Inter~f!diate. I~P'l!~ 889,990 3,031,693 28,785,868 6,124,418 2,923,024 583,620 28,447,601 3,567,048 74,353,262

Ouput_ 2,128,681 6,656,157 92,152,048 14,915,553 6,993,656 2,354,114 56,935,704 13,746,171

1/0 Ratio 0.4181 0.4555 0.3124 0.4106 0.4180 0.2479 0.4996 0.2595

Page 1 of 1

Exhibit3

Source:

941,044,877 $ $ 269,530,101 28.6%

4,350,054 $ 5,316,490 $ 966,436 22.2%

321,820,878 $ 406,256,203 $ 84,435,325 26.2%

$ 1,381,638,286 $ 1,687,240,561 $ 305,602,275 22.1%

$ 5,550,976 26.9%

PREPA: Rate Design Model- Proposed Rates

Proposed Revenues by Class May 27, 2016

No. CEPR-AP-2015-0001

H 000001

Page 1 of 1

Exhibit4

Eight Sectors Transaction Matrix with PREP A's Proposed Tariff Rates Increases

($thousands)

Electricity Mining& Wholesale & Hospitals & & Irrigation Other Total

Agriculture Construction Manufacture Retail Trade Health Serv Serv Services Government Intermediate

Agriculture 357,805 41,461 918,559 51,857 4,200 61 203,026 40,624 1,617,593

Mining & Construction 7,196 80,702 716,221 307,442 64,383 1,433,170 73,292 2,682,406

Manufacturing 340,202 1,591,546 19,722,328 801,313 1,137,327 312,320 4,520,069 1,606,847 30,031,952

Wholesale & Retail Trade 153,090 194,928 5,201,743 268,781 1,053 1,362,585 306,350 7,488,530

Hospitals & Health Serv. 4 1 8 384,304 605 9,052 337,663 731,637 Electricity & Irrigation Serv. I 6,031 41,829 711,609 451,058 40,488 218,094 598,984 183,970 2,252,065

Other Services 126,173 1,014,027 1,554,122 4,539,432 1,026,312 51,487 20,054,734 976,664 29,242,951

Government 1 589 74,772 109,184 55,006 4,562 374,472 80,688 699,273

Intermediat~_ ~-~P1lts 891,086 3,039,269 28,933,767 6,206,116 2,930,357 583,620 28,556,092 3,606,098 74,746,407

Ouput ~-" 2,128,681 6,656,157 92,152,048 14,915,553 6,993,656 2,354,114 56,935,704 13,746,171

1/0 Ratio 0.4186 0.4566 0.3140 0.4161 0.4190 0.2479 0.5015 0.2623

Total Increase in Costs 1,096 7,576 147,899 81,698 7,333 108,491 39,050 393,145

% Increase in Costs 0.12% 0.25% 0.51 o/o 1.33% 0.25% 0.00% 0.38% 1.09% 0.53%

Page 1 of 1

ExhibitS

It is reasonable to think that an increase of 0.53% in overall production costs shall have consequences over production and, in consequence over real GNP in Puerto Rico. To simulate this impact, a forecasting econometric equation was constructed and estimated through the method of Ordinary Least Squares (OLS). The forecasting equation is defined as follows:

Where:

GNPPRt Real GNP in Puerto Rico

Qualitative variable; 0 if fiscal year 2001 to 2006, 1 for fiscal years 2007 to 2017.

Real GDP in the US

Average preferential interest rate in year t

PRICEKWHt Average price of electricity, cents per Kwh

AR(1)

~0

~i

Autoregressive factor, as computed by the Cochrane-Orcutt method

Random error

Intercept

Computed regression coefficient

Data used to estimate the equation is reported on Table S-1.

Page 1 of4

Table 5-1

2000 18,144.8 1,987,336 0.1095 6,487.1 8.60 12,559.7 0

2001 18,723.3 2,386,537 0.1275 6,585.1 8.74 12,682.2 0

2002 19,129.8 2,208,955 0.1155 6,562.6 5.31 12,908.8 0

2003 19,887.3 2,536,250 0.1275 6,702.7 4.42 13,271.1 0

2004 20,260.0 2,613,006 0.1290 6,886.2 4.00 13,773.5 0

2005 20,507.4 3,060,122 0.1492 7,019.6 5.18 14,234.2 0

2006 20,620.3 3,716,082 0.1802 7,35D.6 7.18 14,613.8 0

2007 20,671.6 3,680,390 0.1780 7,261.6 8.25 14,873.7 1

2008 19,601.6 4,362,209 0.2225 7,054.2 6.75 14,830.4 1

2009 18,515.8 4,002,713 0.2162 6,784.2 3.89 14,418.7 1

2010 19,234·.9 4,171,493 0.2169 6,541.8 3.25 14,783.8 1

2011 18,501.4 4,411,213 0.2384 6,431.7 3.25 15,020.6 1

2012 18,112.5 5,052,678 0.2790 6,466.2 3.25 15,354.6 1

2013 18,221.2 4,850,817 0.2662 6,457.6 3.25 15,612.2 1

2014 17,560.9 4,463,163 0.2542 6,347.5 3.25 15,982.3 1

2015 17,280.1 3,977,920 0.2302 6,312.4 3.25 16,397.2 1

1

Table 5-2 informs the results obtained from estimating the forecasting equation, while Table 5-3 presents GNP actual values, as well as those fitted or estimated through the equation, and the residual error, or the difference between actual and fitted or estimated GNP annual values. The table also presents the plot of the residuals.

Page 2 of4

Table 5.2

Dependent Variable: GNP _PR Method: Least Squares Date: 10/07/16 Time: 10:24 Sample (adjusted): 2001 2015 Included observations: 15 after adjustments Convergence achieved after 52 iterations

Variable Coefficient Std. Error !-Statistic

c 7699.575 2071.294 3.717278

DMM -268.7273 158.5572 -1.694829 GDP_US -0.053321 0.135147 -0.394543

R 108.0299 28.09465 3.845214 PRICEKWH -1582.936 1746.380 -0.906410

AR(1) 0.765111 0.111979 6.832606

R-squared 0.913014 Mean dependent var Adjusted R-squared 0.864688 S.D. dependent var S.E. of regression 121.2226 Akaike info criterion

Sum squared resid 132254.3 Schwarz criterion

Log likelihood -89.41732 F-statistic Durbin-Watson stat 1.527659 Prob(F-statistic)

Inverted AR Roots .77

Table 5.3

Pro b.

0.0048 0.1243 0.7024 0.0039 0.3883 0.0001

6717.600 329.5452 12.72231 13.00553 18.89287 0.000156

obs Actual Fitted Residual Residual Plot

2001 6585.10 6772.20 -187.103 I* I I 2002 6562.60 6498.78 63.8204 I I * I 2003 6702.70 6625.32 77.3781 I I * I 2004 6886.20 6760.86 125.345 I I * I 2005 7019.60 7029.22 -9.61626 I *I I 2006 7350.60 7223.76 126.843 I I * I 2007 7261.60 7201.22 60.3807 I I * I 2008 7054.20 7028.05 26.1482 I I* I 2009 6784.20 6768.44 15.7622 I I* I 2010 6541.80 6684.11 -142.310 I * I I 2011 6431.70 6520.63 -88.9292 I * I I 2012 6466.20 6390.01 76.1863 I I * I 2013 6457.60 6485.73 -28.1337 I *I I 2014 6347.50 6473.42 -125.922 I * I I 2015 6312.40 6302.25 10.1500 I I* I

Page 3 of4

If it is assumed an increase of $0.042 in the price of the Kwh, other things equal, then the equation forecasts a reduction of $66.2 million, or 1.05% in real GNP for FY2017. This contraction will be additional to the PR Planning Board forecast of 2.0% decline in real GNP for this fiscal year.

Page 4 of 4

Exhibit6

2000 6,487.1 1,150 0.1773

2001 6,585.1 1,141 0.1733

2002 6,562.6 1,144 0.1743

2003 6,702.7 1,175 0.1753

2004 6,886.2 1,187 0.1724

2005 7,019.6 1,213 0.1728

2006 7,350.6 1,254 0.1706

2007 7,261.6 1,263 0.1739

2008 7,054.2 1,203 0.1705

2009 6,784.2 1,144 0.1686

2010 6,541.8 1,075 0.1643

2011 6,431.7 1,043 0.1622

2012 6,466.2 1,025 0.1585

2013 6,457.6 1,015 0.1572

2014 6,347.5 993 0.1564

2015 6,312.4 984 0.1559

2016 0.1606

0.1673

Employment requirements per million dollars real GNP 167

Real GNP effect of a 11 of 4.2¢ per Kwh -66.20

Change in real GNP _PR -1.05%

Expected reduction in employment -11,075

Page 1 of 1

Exhibit 7

Wholesale and Retail Trade (1/0 Apparel 4.24% 1.33% Matrix) 0.06%

Education & communications 5.10% 0.38% Other services (1/0 Matrix) 0.02% Wholesale and Retail Trade (1/0

Foods & beverages 22.78% 1.33% Matrix) 0.30% Wholesale and Retail Trade (1/0

Other goods and services 9.79% 1.33% Matrix) 0.13%

Housing & housing services Electricity 2.84% 28.6% Residential Tariff (Navigant) 0.81 o/o

Hospitals & Health Services (I/0 Health services 5.47% 0.25% Matrix) 0.01 o/o Entertainment 3.27% 0.38% Other services (1/0 Matrix) 0.01 o/o Transportation Not included

Total exoected increase in CPI

Page 1 of 1

ExhibitS

million of Kwh

Fiscal Year

2000-01 ~-- "~ ~ ... -·

2001-02

PREP A Total

18,723.4

19,129.7

~O_Q?~Q~ ... 19,8§7.4 2003-04

2004-05

2005-06

2006-07

2009-10

2010-11

2011-12

2012-13

2013-14

2014-15

2015-16

Source: PREPA

20,259.9 .

20,~QZ~4. ... 20,620.3

20,671.7

19,601.4

:l?,~l5.8 ····-19,234.9

18,501.4

18,112.5

18,221.2

17,~60.9

1J,.2?Q.J 17,349.1

Page 1 of 2

21,000.0

20,000.0

19,000.0

..c:

~ '5 " 18,000.0

~ ~

17,000.0

16,000.0

15,000.0

PREPA's Net Production of Electricity

Fiscal Years

Page 2 of 2

Exhibit 9

It is reasonable to expect that an increase in electricity tariff rates of the magnitude of 26.5% shall have some effect on the quantity demanded of electricity. To evaluate this effect, it was estimated a demand equation for electricity. The equation is defined as follows:

Where:

DMMt:

PREP A's net production of electricity in year t

Qualitative variable; 0 if fiscal year 2001 to 2006, 1 for fiscal years 2007 to 2017

PRICEKWHt Average price of electricity, cents per Kwh in year t

Real GNP in Puerto Rico in year t

PDPt Total population in Puerto Rico in year t

Random error

f3o Intercept

Computed regression coefficient

Data used to estimate the equation is informed in Table 9-1

Page 1 of 3

2000 18,144.8 0.1095 6,487.1 3,808 0 2001 18,723.3 0.1275 6,585.1 3,828 0 2002 19,129.8 0.1155 6,562.6 3,549 0 2003 19,887.3 0.1275 6,702.7 3,869 0 2004 20,260.0 0.1290 6,886.2 3,887 0 2005 20,507.4 0.1492 7,019.6 3,903 0 2006 20,620.3 0.1802 7,350.6 3,813 0 2007 20,671.6 0.1780 7,261.6 3,794 1 2008 19,601.6 0.2225 7,054.2 3,772 1 2009 18,515.8 0.2162 6,784.2 3,751 1 2010 19,234.9 0.2169 6,541.8 3,731 1 2011 18,501.4 0.2384 6,431.7 3,700 1 2012 18,112.5 0.2790 6,466.2 3,657 1 2013 18,221.2 0.2662 6,457.6 3,614 1 2014 17,560.9 0.2542 6,347.5 3,564 1

2015 17 280.1 12.4 OS 1

Table 9.2 informs the results obtained from estimating the demand equation, while Table 9.3 presents PREP A's actual values for net production, as well as those fitted or estimated through the equation, and the residual error, or the difference between actual and fitted or estimated GNP annual values. Table 9-3 also presents the plot of the residuals.

Page 2 of 3

Table 9.2

Dependent Variable: ELECT_CF Method: Least Squares Date: 10/07/16 Time: 14:05 Sample (adjusted): 2001 2015 Included observations: 15 after adjustments

Variable Coefficient Std. Error !-Statistic

c 1989.421 1113.599 1.786479 DMM 195.9176 104.9654 1.866496

PRICE KWH -7663.317 969.8034 -7.901928 GNP_PR 2.082373 0.113758 18.30523

POP 1.221540 0.335938 3.636205

R-squared 0.993466 Mean dependent var Adjusted R-squared 0.990852 S.D. dependent var S.E. of regression 101.1640 Akaike info criterion Sum squared resid 102341.5 Schwarz criterion Log likelihood -87.49423 F-statistic Durbin-Watson stat 1.111780 Prob(F-statistic)

Table 9.3

Pro b.

0.1043 0.0915 0.0000 0.0000 0.0046

19120.28 1057.701 12.33256 12.56858 380.0976 0.000000

obs Actual Fitted Residual Residual Plot 2001 19134.8 19401.0 -266.205 I* I I 2002 19141.9 19105.3 36.5832 I I* . I 2003 19793.9 19696.0 97.8711 I I * I 2004 20184.3 20088.6 95.7061 I I * I 2005 20292.2 20231.1 61.0905 I I *. I 2006 20547.9 20572.9 -25.0461 I *I I 2007 20548.9 20577.1 -28.2938 I *I I 2008 19732.1 19777.4 -45.2937 I *I I 2009 19242.0 19237.8 4.24275 I * I 2010 18744.7 18703.2 41.5414 I I* . I 2011 18305.7 18271.3 34.3824 I I* . I 2012 17962.2 17979.5 -17.2920 I *I I 2013 18004.7 18007.1 -2.45473 I • I 2014 17836.2 17808.7 27.4494 I I* I 2015 17332.8 17347.1 -14.2816 I *I I

Page 3 of 3

\

1

2

3 A. We are Tom Sanzillo and Cathy Kunkel. We are jointly sponsoring this testimony.

4 Tom Sanzillo is the Director of Finance for the Institute for Energy Economics and Financial

5 Analysis. His business address is 3430 Rocky River Drive, Cleveland, OH 44111.

6 Cathy Kunkel is an Energy Analyst with the Institute for Energy Economics and Financial

7 Analysis. Her business address is 3430 Rocky River Drive, Cleveland, OH 44111.

8 Q. PLEASE STATE YOUR QUALIFICATIONS.

9 Tom Sanzillo is the author of several studies on coal plants, rate impacts, credit analyses, and

10 public and private financial structures for the coal industry. He has testified as an expert witness,

11 taught energy-industry finance training sessions, and is quoted fi·equently by the media. Sanzillo

12 has 17 years of experience with the City and the State of New Y ark in various senior financial and

13 policy management positions. He is a fmmer first deputy comptroller for the State ofNew York,

14 where he oversaw the finances of 1,300 units of local government, the annual management of

15 44,000 government contracts, and where he had oversight of over $200 billion in state and local

16 municipal bond programs and a $156 billion pension fund.

17 Sanzillo recently contributed a chapter to the Oxford Handbook ofNew York State Government

18 and Politics on the New York State Comptroller's Office.

19 Sanzillo has a bachelor' s degree fi·om the University of Califomia in politics.

Page 1

20 Cathy Kunkel has co-authored numerous reports for the Institute for Energy Economics and

21 Financial Analysis related to utility regulation, electricity markets, mergers and acquisitions, and

22 coal plant finances. Previously she was a Senior Research Associate in the Electricity Markets and

23 Policy group at Lawrence Berkeley National Laboratory. She has been an expeti witness in eight

24 West Virginia Public Service Commission proceedings regarding resource planning and energy

25 efficiency. She has also participated in hearings before the Puerto Rico Energy Commission in its

26 Integrated Resource Plan proceeding.

27 Kunkel graduated from Princeton University with a B.A. in physics and from Cambridge

28 University with a Certificate of Advanced Study from the Depatiment of Applied Mathematics

29 and Theoretical Physics.

30 Our resumes are attached as Exhibits 1 and 2.

31 Q. ON WHOSE BEHALF ARE YOU TESTIFYING?

32 A. We are testifYing on behalf of ICSE-PR, the Institute for Competitiveness and Sustainable

33 Economy.

34 Q. PLEASE SUMMARIZE YOUR TESTIMONY.

35 A. Our testimony addresses the reasonableness of PREP A's proposed rates, from the perspective

36 of their affordability; the reasonableness of the budget assumptions embedded in the rates; and the

37 likelihood that the proposed rates will be adequate to support PREP A's re-entry to the credit

38 markets. We find that the proposed rates are excessively high when benclnnarked against other

39 U.S. jurisdictions and unjustly biased in that they force industrial and commercial ratepayers to

40 subsidize other customer classes. However, we also find that the budget assumptions embedded in

Page 2

41 the rates are umealistic: if they are not met, PREP A's debt service coverage ratio (a key credit

42 metric) will likely fall below what PREP A's consultants believe is necessary for re-entry to the

43 bond markets. This will result in upward pressure on rates in future years, exacerbating our

44 concerns about affordability. Our conclusion is that PREP A's ratepayers cannot supp01i the level

45 oflegacy debt (inclusive of the previously approved Transition Charge) embedded in the proposed

46 rates. Although we are conscious that debt restructuring is not pari of this proceeding, the overall

47 effect of the proposed rate increases points to the need for further debt renegotiation.

48 Additionally, our testimony addresses the need for close Commission oversight over PREP A's

49 expenditures and makes recommendations regarding PREP A's arguments for a formula rate-

50 making mechanism.

51 Thirdly, our testimony finds that the proposed rate design does not give customers, particularly

52 commercial and industrial classes, the :flexibility to lower their own energy costs and to expand

53 the use of renewable energy generation in Puerto Rico. We recommend that open access in

54 transmission and distribution be implemented and that the industrial and commercial tariff designs

55 be weighted less heavily towards demand charges.

56 Finally, we attach an addendum (Exhibit 3) that summarizes our direct responses to some of the

57 Commission's questions to intervenors provided in the September 27,2016 resolution.

58 I. Proposed rates are unreasonable in comparison with other utilities

59 Q. HOW DO PREPA'S PROPOSED RATES COMPARE AGAINST RATES IN THE

60 MAINLAND U.S.?

61 A. The rates proposed by PREP A ar·e unaffordable. The testimony of PREP A witness Kaufman

62 benchmarked PREP A against mainland U.S. utilities in tenus of operating costs and operating

Page 3

63 revenues, but did not look at affordability metrics. The following table benchmarks PREP A

64 against other states with comparably high electricity rates. Under PREP A's projected FY 2017

65 rates, Puerto Rico will pay twice the average price of electricity in the U.S. in 2014. Puerto Ricans

66 have, on average one third the median income of households in those states with high electricity

67 rates. In addition, and unlike any of the following states with comparably high electricity rates

68 except Alaska, Puerto Rico had negative average GDP growth for the period 2011-2015 1 and

69 economic contraction is expected to continue, with GDP projected to decline by 2.0% over the

70 next year. 2

71 Additionally, Puerto Rico's economy relies on manufacturing to a far greater extent than any other

72 state with comparably high electricity rates and the contribution of manufacturing to GDP is nearly

73 four times the U.S. average. The large contribution that this sector makes to Puerto Rico's economy

7 4 implies that economic losses in this sector will have a bigger impact on the overall Puerto Rican

75 economy than comparable manufacturing losses would have in other U.S. states. Yet, in other

76 states with high electricity rates, the industrial rate is considerably below the average rate. 3. And,

77 as shown by the testimony of Dr. Ramon Cao on behalf of ICSE-PR, PREP A is proposing a rate

78 increase that will lead to further contraction in Puerto Rico's economy.

79

1 1n fact, aside from a small increase from 2011-2012, Puerto Rico's real GDP has declined since 2006. (Government Development Bank, Gross National Product in 1954 dollars, http:/ /www.gdb-pur.com/economy/statisticalappendix.html) 1 PREPA's 2014 Audit highlights the disparity between United States GDP and Puerto Rico's. http://www.aeepr.com/INVESTORS/DOCS/Financial%201nformation/Annuai%20Reports/Financiai%20Statements, %20Required%20Supplementary%201nformation%20and%20Supplementai%20Schedules%202014.pdf, p. 26. 3 Industrial electric rates for mainland U.S. available from U.S. Energy Information Administration, www.eia.gov/electricity/data/browser/. The industrial rate for Puerto Rico for FY 2017 can be estimated by dividing total revenues for class GSP ($920 million) by total sales (4,510 GWh), as provided in Schedule H.

Page4

80

81

82 Table 1: Comparison of Puerto Rico's proposed FY 2017 rates with recent (2014) electric 83 rates in U.S. states and other economic indicators.

Electricity Real GDP

Fraction of GDP State Median (Annual Average

Rates from (cents/kwh) 4 Incomes5 2014 2011 through

manufacturing7 2015)6

Hawaii 33.43 $71,223 1.2% 1.93% Puerto Rico 20.108 $19,183 (0. 5%)9 47.6% Alaska 17.46 $67,629 (1.0)% 2.44% Connecticut 17.05 $70,161 0.4% 10.68% New York 16.25 $54,310 1.4% 5.18% U.S. Total10 10.44 $53,657 2.0% 12%

84

85 Q. WHAT FACTORS ARE CONTRIBUTING TO PREP A'S HIGH ELECTRICITY RATES?

86 A. Fuel and purchased power costs historically (FY 2012- FY 2014) have comprised 75-80% of

87 PREP A's total rate, in part because of the failure to update the base rate for several decades. Going

88 forward, PREP A projects that fuel and purchased power costs will be approximately half of the

89 total rate. In addition, PREP A's proposed rates contain an unreasonably high level of debt and debt

90 service related costs. 11 In FY 2017, Schedule A-1 REV lists "Debt Service (Principal & Interest)"

4 U.S. Energy Information Administration, "State Electricity Profiles", 2014. http://www.eia.gov/electricity/state/ 5 For all states see: http://www.advisorperspectives.com/dshort/updates/Household-lncomes-by-State and the Commonwealth see: https :/ /www.census.gov I content/ dam/Census/library /pu blications/2014/acs/acsbrl3-02.pdf 6 U.S. Bureau of Economic Analysis, "Real GDP by State (millions of chained 2009 dollars)", June 2016. 7 National Association of Manufacturers, "State Manufacturing Data," March 2016 (http://www.nam.org/Data-andReports/State-Manufacturing-Data/State-Manufacturing-Data/2015-State-Manufacturing-Data-Table!) and Government Development Bank of Puerto Rico, "Puerto Rico Fact Sheet", March 2016 (http://www.gdbpur.com/economy/documents/PREconomicFactSheet-March2016.pdf) 8 FY 2017 rate from Schedule F. The FY 2014 rate for Puerto Rico was 26.4 cents/kWh 9 Government Development Bank, "Gross National Product in 1954 dollars", http://www.gdbpur .com/economy /statistica 1-a ppendix.html 10 Of the 50 states Mississippi has the lowest median income in 2014: $35,521. 11 Exhibit 14.01, Tab D2A compares the outstanding value of PREP A's bond indebtedness to a market valuation of the bonds. The outstanding value is $8.1 billion and the market value is $5.3 billion. Cell S224 shows that the market value is 65.3% of the outstanding value of the bonds. PREP A has presented to the Commission, and the Commission

Page 5

91 of $314.3 million, "Debt Service for Securitization" of $394 million, "Gross-up for Collections

92 Lag and Uncollectible Revenue" (specific to the securitization charge) of $109 million, and

93 "Capital Expenditure" of $337 million. 12 In total, debt and debt service related costs amount to

94 $1.154 billion, or 6.7 cents/kWh, one-third of the proposed total rate.

95 Q. HOW DOES THIS LEVEL OF DEBT COMPARE TO OTHER PUBLIC POWER ENTITIES

96 IN THE UNITED STATES?

97 A. Table 2 benchmarks PREP A's debt and debt service related costs against major public power

98 utilities in the mainland. PREP A has- by far- the highest propmiion of its total rate going towards

99 debt and debt service related expenses. (This would still be true even if we did not include revenue-

100 financed capital expenditures in this category).

101 The compmison to LIP A (the Long Island Power Authority), which has the second largest debt

102 service shown in the table, is instructive. In 1998 the Long Island Power Authority (LIP A) issued

103 $7 billion in long term bonds to pay costs incuned with the decommissioning of the Shoreham

104 Nuclem Power Plant13. In 2013 the State of New York created the Utility Debt Securitization

105 Authority (UDSA). The UDSA was created to absorb a substantial portion of the remaining

106 liability. Since 2013 the UDSA has issued $3 billion in long term debt and plans another $1 billion.

approved a securitization arrangement where 85% of a portion of the bond indebtedness is refinanced and the remaining PREP A legacy indebtedness is paid at 100% of outstanding value. There is no attempt to reconcile the sizable difference between the market value of the debt and what PREP A ratepayers are being asked to underwrite. 12 We include "capital expenditure" in the category of debt and debt service related costs, because if PREP A had access to the capital markets, this would be financed with debt. 13 Long Island Power Authority and Subsidiaries, Consolidated Financial Statements 2000 and 2001, http:ljwww.lipower.org/pdfs/company/investor/lipa financials2001.pdf, p. 5.

Page 6

107 LIP A, now a separate operating entity retains an estimated $2.2 billion14 in existing long te1m debt.

108 Some of this debt is legacy debt and some of it has been used to fund new capital needs.

109 There are two noteworthy similarities between PREP A and LIP A. First, each has adopted a very

110 similar corporate structure to facilitate management of its long term indebtedness. Second, each

111 has incurred substantial debt for which there is no underlying specific asset that generates revenue

112 to pay the debt.

113 The fundamental difference in the two utilities is their economic enviromnent. PREP A is being

114 asked to carry over $8 billion in long te1m debt in a territory with a household median income of

115 $19,183. Nassau County and Suffolk County (LIP A's service area) are carrying indebtedness of

116 approximately $4-$5 billion in areas with median incomes of $98,401 15 and $88,323 16 ,

117 respectively.

118 Table 2: PREP A's debt service compared to other large public power entities.

Revenues Debt DS as% of Debt DS as% of Service Generation Rate

(R) (DS) Revenue Service Rate

$billions $billions Percentage Billion kwh Cents/ Cents/

Percentage kwh kwh

PREP A 2.96 1.15 38% 17.27 6.7 20.1 33% LIPA17 3.4 0.6 18% 20.4 2.9 18.0 16% Santee 0.28

16% 1.0 7.3 14% Cooper18 1.8 26.5

14 Long Island Power Authority, 2016 Approved Operating Budget, 2016 Approved Capital Budget, 2017 and 2018 Projected Operating and Capital Budget, http://www.lipower.org/profile/2016%20APPROVED%20BUDGET.pdf,

page b-2" 15 U.S. Census, "Quickfacts: Nassau County, NY", http://www.census.gov/quickfacts/table/SB0020212/36059 16 U.S. Census, "Quickfacts: Suffolk County, NY", http://www.census.gov/quickfacts/table/RHI105210/36103 11 Long Island Power Authority (LIPA) http://www.lipower.org/profile/2016%20APPROVED%20BUDGET.pdf, p. 5 rates, p. a-2 Revenue and Generation, p. A, Debt Service. " Santee Cooper, Annual Report 2015, https:ljwww.santeecooper.com/pdfs/about-santee-cooper/201Sar/201SAR FINAL. pdf. Revenues and Interest, p. 13 and Generation p. 6.

Page 7

AMP19 1.1 0.14 13% 14.0 1.0 7.9 13% LAWPD20 3.3 0.46 14% 25.4 1.8 11.1 16% Salt River 0.30

10% 0.9 11.3 8% Project21 3 33.6 CPS 0.33

13% 1.0 17.0 6% Energy 22 2.6 33.2

119

120 Q. IS PREP A PROPOSING TO ALLOCATE COSTS EQUITABLY ACROSS CUSTOMER 121 CLASSES?

122 A. No. The proposed PREP A rates disadvantage commercial and industrial customers because

123 these classes are excessively bearing some of the cost of serving other rate classes. As shown in

124 Exhibit G-3, industrial customers on tariffs GSP, GST, TOUP and TOUT will pay 26.5%, 35%,

125 14.6% and 34.6% more, respectively, than would be required to achieve the results of the

126 Embedded Cost of Service Study. The graph at line 421 in the testimony of PREP A witnesses

127 Zarumba and Granovsky shows that, to meet the Embedded Cost of Service Study, the commercial

128 class should see a 6.1% rate increase and the industrial class a 1.4% rate increase; instead PREP A

129 is proposing a 22.1% rate increase for commercial customers and 26.2% for industrial customers.

130 PREP A Witnesses Zarumba and Granovsky state that "equitable allocation of the revenue

131 requirement" among customer classes is one of the objectives of PREP A's new rate design

132 (Exhibit 4.0, lines 45-62), but that this cannot be achieved immediately in this case. There is no

133 specific target date established nor phase-in plan in the rate design for achievement of this

19 American Municipal Power Inc., Consolidated Financial Statements and Supplementary Information 2015 and 2014, http://www .a m ppa rtne rs. o rg/ docs/ de fa u It -sou rce/i nvesto rs/fi na nc ia I-re po rts/20 15/ a m p co nso I i dated fs 2 015, p df?sfv rs n-2, p. 5. 20 los Angeles Power and Water Department (LAPWD), Power System Revenue Bonds 2016 Series B, http://emma.msrb.org/ER970367-ER758976-ER1160417.pdf, 6/30/15, Revenues and Debt Service, p. 52 and Generation page 54. 21 Salt River Project, Electric System Revenue Bonds, 2015 Series A: http://emma.msrb.org/EP863182-ER690004-ER1091632.pdf. Revenues and Generation, p. 31, Rates p. 32, Debt Service, p. 38 n CPS Energy, FY 2015 Annual Report:

https://www.cpsenergy.com/content/dam/corporate/en/Documents/Finance/FY 2015 Annual Report.pdf, Three Year

Highlights Unaudited, electronic page 17.

Page 8

134 objective. The fact the proposed rate increase is so high means that residential rates would have to

135 increase by more than 60% in order to align with the cost of service study (Exhibit 4.0, lines 420-

136 421). In order to avoid this level of residential rate shock, PREP A is proposing rates that have the

13 7 residential class subsidized by the commercial and industrial classes.

138 This is simply another indication that the overall rates are unaffordable. If, as Zarumba and

139 Granovsky indicate, moving towards equitable allocation between classes is one of PREP A's

140 objectives, then PREP A will be raising the residential rate even faster than the 38% overall rate23

141 increase projected fi·om FY 2016 through FY 2021.24

142 Although we are conscious that debt restructuring is not part of this proceeding, the overall effect

143 of the proposed rate increases is unsustainable, pointing to the need for finiher debt renegotiation.

144 The PREP A debt urgently needs to be renegotiated in order to bring the overall level of rates down,

145 so that each class can afford to pay its own costs, without cross-subsidization, in accordance with

146 sound rate-making practices.

147 II. Rates are likely to go higher than what PREP A projects

148 Q. WHAT LEVEL OF RATE INCREASE IS CURRENTLY PROPOSED BY PREP A?

149 A. PREP A is proposing an overall rate of20.1 cents/kWh in FY 2017, compared with actual rates

150 in the last fiscal year of 18.52 cents/kWh.25 This rate includes the Transition Charge, an increase

151 in the base rate, the implementation of various savings initiatives and a projected decrease in fuel

23 The actual rate in FY 2016 was 18.52 cents/kWh, according to PREP A's unaudited June 2016 monthly report. We use this number as the FY 2016 rate, as opposed to the 17.79 cents/kWh rate presented in Schedule F for FY 2016. 24 1n response to the Commission's 4th Request for Information (questions CEPR-01-03 and CEPR-01-04), witness Zarumba states that it will likely take several rate requests before an equitable allocation is achieved, but no specific timeframe is proposed. 25 PREPA Monthly Report to the Governing Board, June 2016, http://www.aeepr.com/INVESTORS/DOCS/Financial%201nformation/Monthly%20Reports/2016/June%202016.pdf,

p. 26

Page 9

152 costs (See Schedule A-1 REV). By FY 2021, PREP A projects rates increasing to 25.6 cents/kWh,

153 a 38% increase above cmTent rates. PREP A projects stable rates at 24-25 cents/kWh through FY

154 2030.

155 We note that PREP A's actual FY 2016 rate of 18.52 cents/kWh is higher than the estimated FY

156 2016 rate of 17.79 cents/kWh shown in Schedule F-1. PREP A's underestimate of the FY 2016 rate

15 7 is not explained and raises questions about the validity of FY 2017 cost assumptions.

158 Q. DO YOU THINK PREPA'S PROJECTIONS OF FUTURE RATE INCREASES ARE

159 ACCURATE?

160 A. No. We think it likely that rates will go even higher than PREP A projects in Schedule F-1

161 because revenues will not be as high as expected and operational expenses will be higher than

162 budgeted. Either or both of these outcomes will increase pressure to raise rates in future years.

163 Additionally, if, as we believe, PREP A has underestimated its fuel expenditures for FY 2017, this

164 will lead to quarterly rate increases in FY 2017.

165 This will only worsen the problems of affordability and overall economic impact of the proposed

166 rate stmcture described in Section I, above.

167 Q. PLEASE EXPLAIN WHY THERE IS A RISK THAT REVENUES WILL BE LOWER THAN

168 ANTICIPATED.

169 A. We believe that PREP A's long-term sales forecast is too high. PREP A forecasts that electricity

170 consmnption in Pumio Rico will remain essentially flat through 2030, as shown in the following

171 graph. 26

26 Schedule F-1.

Page 10

172

173

174

175 176 GRAPH: PREP A Actual and projected sales of electricity with IEEFA adjustment to 177 PREP A outlook (1994-2034)

178

Actual and projected sales 25,000 -,-~~~~~~~~~~~~~~~~~~~~~~-

20,000 +----_--::: ,_~-"""""""";,;;;;;;::::::=-----------~

:2 ~ ---$ 15,000 +-"""'-'--~~~~~~~~~~~-=-...__-=--~~~~~-.£'. ~ ~ 10,000 +-~~~~~~~~~~~~~~~~~~~~~~~~-

"' V)

5,000 +-~~~~~~~~~~~~~~~~~~~~~~-

-Historic Sales ~PREPA sales forecast -Continuation of declining trend

179 PREP A has faced declining electricity consumption for the past several years.27 PREP A's sales

180 declined 16% from 2007 through 2015, from 20.6 billion kWh to 17.3 billion kWh. Sales were flat

181 at 17.3 billionkWhinFY2016.

182 Q. HOW DOES PREP A JUSTIFY ITS SALES FORECAST?

183 A. In response to discovery (ICSE-PR request 26), PREP A states that electricity sales have

184 historically been con·elated to Puetto Rico's GDP. The economic forecast fi·om Inter-American

185 University Global Insight shows GDP starting to increase in FY 2017. PREPA states that FY

' 7 Declining electricity sales are cited as a challenge for the Authority in its 2014 Audited financial statement. http://www.aeepr.com/INVESTORS/DOCS/Financial%201nformation/Annuai%20Reports/Financiai%20Statements, %20Req u ired%20Supp lementa ry%20information%20a nd%20Su pplem entai%20Sch ed u I es%202014. pdf, p. 7.

Page 11

186 2016's 0.4% increase in electricity sales over FY 2015 further supports the idea that Puerto Rico's

187 economy is beginning to expand. Additionally, PREP A notes that its sales forecast has been

188 accurate for the last two years. ("PREP A's forecast shows a growth of0.2% each year, and this is

189 reasonable considering the accuracy of the last two projections and the consumption behavior in

190 FY 2016").

191 Q. DOES PREPA'S ANALYSIS CONSIDERTHE IMPACT OF REDUCED ELECTRICITY

192 PRICES ON FY 2016 SALES?

193 A. PREP A's forecasting methodology document provided in response to Commission request

194 CEPR-AH-1-05 (Attachment 8) indicates that the electricity price is included in the forecasting

195 model. Given that PREP A's rates declined 22% from FY 2015 to FY 201628, this should not be

196 discounted as a driver of PREP A's increased FY 2016 sales.

197 Q. WHAT ASSUMPTIONS UNDERL Y PREP A'S FUTURE SALES FORECAST?

198 A. PREPA's forecasting methodology document describes the sales forecast as based on

199 underlying Puerto Rico economic indicators and the electricity price. The document states that

200 these economic indicators were obtained from the projections oflnter-American University Global

201 Insight.

202 The response to ICSE-PR Question 26 indicates that IAUGI's projection shows GDP bottoming

203 out in FY 2016, in contrast to the Puerto Rico Planning Board's forecast of continued GDP decline

204 in FY 2017. 29 Previously, PREP A has used the lower of the IAUGI, Planning Board, and

28 PREPA Monthly Report to the Governing Board, June 2016, p. 5, http://www.aeepr.com/INVESTORS/DOCS/Financiai%20information/Monthly%20Reports/2016/June%202016.pdf " D. Costa, "Planning Board: Puerto Rico Economy to Drop 2% in Fiscal 2017," Caribbean Business, May 6, 2016, http:/ I cb. pr /planning-boa rd-puerto-rico-economy-to-d rap-2-in-fisc a 1-2017/

Page 12

205 Advantage Business Consulting forecasts in creating its revenue forecasts. 30 That methodology.

206 has not been used in this case.

207 Q. WHY DO YOU EXPECT THE DOWNWARD TREND IN ENERGY SALES TO

208 CONTINUE? WHAT IMP ACT WOULD TI-IIS HAVE ON RATES?

209 A. PREP A's past statements and data show that Puerto Rico's economic growth and the overall

210 level of electric rates are both impmtant drivers of changes in electricity consumption. Based on

211 the Puerto Rico Planning Board's forecast of continuing GDP decline in FY 2017, the Pue1to Rico

212 Fiscal Plan's projection of real GDP decline through 202631 and PREP A's plan to raise rates 38%

213 by FY 2021, we think is likely that PREP A's long-te1m forecast of flat electricity consumption is

214 too high. We have adjusted the above graph to create an illustrative scenario based upon a

215 continuation of PREP A's declining trend32

216 Lower-than-forecast electricity consumption results in the spread of costs of electricity production

217 over a smaller sales base (kWh sold). Because a large fraction of PREP A's production costs are

218 fixed costs, in part stemming from PREP A's high debt levels, the net effect is upward pressure on

219 rates.

220 Q. WHY IS THERE A RISK THAT PREPA'S OPERATIONAL EXPENSES COULD BE

221 HIGHER THAN FORECAST?

30 PREPA Power Revenue Bonds, Series 2013A, http://www.aeepr.com/INVESTORS/DOCS/Financial%201nformation/Officiai%20Statement/PREPA%20Revenue%2 0Bonds%20Series%202013A.pdf, p. 39-40. 31 Commonwealth of Puerto Rico Fiscal Plan, October 14, 2016 (http://www. fortaleza. pr.gov I sites/ de fa u lt/fi les/16 .10.14 %20Fiscai%20Pian%20vFi na I. pdf), p. 82. 321n its IRP Order, the Commission faulted PREP A for failing to develop a load forecast sensitivity that was significantly lower than its baseline load forecast, noting that "a substantially lower forecast could be justified, at the least, by the recent loss of Puerto Rico" (IRP Order, September 23, 2016 at paragraph 133).

Page 13

222 A. We have identified four risks: the risk of (a) slippage in PREP A's proposed savings initiatives;

223 (b) higher than anticipated fuel costs; (c) funding capital expenditures tln·ough revenues for the

224 next several yeaTs; (d) increases to the transition charge.

225 Q. PLEASE EXPLAIN HOW PREPA'S PROPOSED SAVINGS INITIATIVES ARE

226 RELEVANT TO ITS RATE FORECAST.

227 A. PREP A has proposed and is implementing numerous savings initiatives. These initiatives,

228 according to the testimony of Miranda, Sales and Sosa, have "already achieved approximately

229 $165 million in one-time cash savings and approximately $200 million in recurring annual

230 savings," and PREP A "forecasts to save an incremental $120 million ofrecuning annual savings

231 before 2019" (lines 180-183). If these savings initiatives do not materialize as forecast, rates will

232 need to be raised to compensate.

233 Q. WHY DO YOU BELIEVE THAT PREP A IS UNLIKEL YTO MEETITS SAVINGS GOALS?

234 A. The Commission has raised several issues33 with PREP A that relate to the organizational

235 preparedness of the Authority34 to cany out the administrative and budget actions required to

236 successfully implement its reorganization.

237 In PREP A's third response to the Commission it has acknowledged that PREP A's organizational

238 capacity is uneven with "varying degrees of preparedness for each improvement areas."35 Our

239 concern about organizational capacity is exacerbated by the departure of Sonia Miranda, the only

33 The Commission's August 2, 2016 resolution and order provides the background of the Commissions efforts to secure details of the organizational issues confronting PREP A. The Commission has been probing witnesses Miranda and Donahue concerning their unspecified statements regarding: 1) a history of poor accountability in PREP A; 2) political interference and 3) staff capacity. This line of inquiry goes directly to the point that the savings initiatives may not materialize and that the program reforms may not solve PREP A's long standing problems. 34 The Commission's recent order in the IRP case also raises significant concerns regarding PREP A's organizational preparedness. (IRP order, September 23, 2016 paragraph 13). 35 CEPR-SH-001-009(b)

Page 14

240 PREP A employee who sponsored the original panel testimony on the savings initiatives (Exhibit

241 3.0).

242 Recent PREP A management decisions also call into question its ability to control costs. For

243 example, PREP A entered into a large number of above-market contracts for solar fi"om 2010-

244 2013 36 . Additionally, in selecting contractors for its bond restructuring, where contractors are

245 being paid tens of millions of dollars, PREP A failed to use competitive bidding.37

246 In addition, PREP A has not presented its savings initiatives in a transparent and consistent marrner

247 that would provide confidence in PREP A's ability to meet the targets.

248 Q. HOW DOES PREP A'S APPROACH TO ITS SAVINGS INITIATIVES CONTRAST WITH

249 BEST PRACTICES?

250 A. Typically when public agencies are involved in a large series of initiatives to bring a budget

251 into balance, the actions are tied together in what is sometimes called a "program to eliminate the

252 gap" (PEG). 38 The program creates a uniform system of accountability that identifies specific

253 budget and organizational initiatives and how the initiatives save money or generate additional

254 revenue, sets out financial targets, creates standards of accounting for the measurement of the

255 benefit, assesses related risk, establishes timelines and benchmarks to measure progress toward

256 objectives and assigns responsibility to administrators for achievement of objectives and conective

257 action plans.

36 IRP Order, September 23, 2016, paragraphs 179 and 184. 37 Restructuring Order (June 21, 2016) paragraphs 258-262. 38 See, for example: http://policyatlas.org/wiki/Program_to_Eiiminate_Gap_proceduresjPEG)

Page 15

258 On a budget-wide basis the PEG initiatives are integrated into the budget process of specific units,

259 agencies of government and into executive level budget documents. A cmeful tracking 39 is

260 maintained to monitor agency progress and to take action when early warning signs show slippage

261 in meeting perfonnance objectives.

262 In contrast, PREP A's presentation of savings initiatives and revenue-producing actions identified

263 in the rate docket do not present a unifonn system that is transpment, easily understandable, or

264 usable for the kind of rigorous budget monitoring that one would expect given the size ofPREPA's

265 budget imbalance.40

266 For example, Schedule F-4 provides information on assumptions underlying PREP A's FY 2017

267 revenue requirement. Under "customer service improvement savings", Schedule F -4 lists changes

268 to reconnection chmges, theft recoveries and reduced T &D losses, which together appem· to total

269 approximately $30 million in FY 2017. Under "other improvement savings", Schedule F-4 appems

270 to show more than $54 million in savings in FY 201 7. Yet Schedule A -2 shows only $23.7 5 million

271 and $24 million in FY 2017 savings in these two categories, respectively.

272 Finally, in its response to the Commission's third request for information (CEPR-SH-001-006),

273 PREP A states that it does not yet have mechanisms in place to compensate or penalize managers

27 4 if improvements are achieved or not achieved, though PREP A plans to malce this "a component of

275 the annual review process" and a "factor detennining futrn·e cmeer advancement opportunities,"

39 Often outside groups with specific budget interests monitor these initiatives as well. See, for example: http:Uwww.cbcny.org/category/tags/program-eliminate-gap

40 Here we contrast the system monitoring for operational expenses with the system monitoring used to insure that revenues are collected and debt service is paid. The debt service system is elaborate with significant attention paid to internal control accounting and revenue disbursement. See entire Testimony of Michael Mace1 Transition Docket, with summary chart at Line 1558.

Page 16

276 though no specifics were provided. This again raises the question of whether there will be

277 appropriate incentives within PREPA's organizational culture for achievement of savings

278 initiatives.

279 Q. YOU ALSO CITED HIGHER THAN ANTICIPATED FUEL COSTS AS A FACTOR THAT

280 COULD PUT UPWARD PRESSURE ON RATES. PLEASE EXPLAIN.

281 A. PREPA's estimated fuel expense for FY 2017 according to Schedule A-1 REV is

282 $655,968,367; according to Schedule A-6 REV, it is $763,695,078. This represents a $1.581 to

283 $1.689 billion reduction in fuel charges from the 2014 audited statement of $2.344 billion. The

284 reduction is carried as a cost savings by PREP A in its presentation to the Commission.41 PREP A

285 then projects fuel costs to rise again in FY 2018, back to approximately the FY 2016 level

286 (Schedule A-6 REV).

287 It is umealistic to expect PREP A to pay only $656-$764 million for fuel in FY 2017. PREP A has

288 recently provided its interim, unaudited Monthly Report for June 2016, the end ofPREPA's Fiscal

289 Year (FY). The unaudited data puts PREP A's FY 2016 fuel costs at $1.210 billion42 based upon

290 consumption of 23,202 barrels43 at an average cost of $52.1744 per barrel.

41 Pam push, Porter and Stathos Direct Testimony Line 475 puts the savings at $1.595 billion. 42 PREPA Monthly Report to the Governing Board, June 2016, http://www.aeepr.com/INVESTORS/DOCS/Financial%201nformation/Monthly%20Reports/2016/June%202016.pdf. 8. Fuel Consumption (Cost- Fiscal Year Total) 43 PREPA Monthly Report to the Governing Board, June 2016, http://www.aeepr.com/INVESTORS/DOCS/Financial%201nformation/Monthly%20Reports/2016/June%202016.pdf, 2. Fuel Consumption (BBL- Fiscal Year Total). 44 PREPA Monthly Report to the Governing Board, June 2016, http://www.aeepr.com/INVESTORS/DOCS/Financial%201nformation/Monthly%20Reports/2016/June%202016.pdf, I. Operations Highlights, A. Production, 4. Average Cost Per Barrel, Fiscal Year

Page 17

291 Schedule E-7 REV projects a 14% decline in barrels of oil consumed in FY 2017 relative to FY

292 2016. Even so, in order for fuel costs to decline fi·om $1.210 billion in FY 2016 to $656 million in

293 FY 2017, a 30%-40% drop in fuel prices would be required. This is not realistic.45

294 We have surveyed recent oil price forecasts from four prominent, independent sources. None of

295 the oil price forecasts anticipate a precipitous drop in the price of oil for 2017. Prices in 2017 are

296 expected to be flat compared to 2016, according to the United States Energy Information Agency

297 ($50.58 per barrel)46, World Banlc ($53 per barrel), International Monetary Fund ($51 per barrel)

298 and The Economist Intelligence Unit (EIU) ($55 per banel)Y Additionally, PREP A's own fuel

299 forecast used in its IRP does not predict any significant decline in PREP A's fuel costs from FY

300 2016 to FY 2017.48

301 In Schedule F, PREPA informed the Commission that it anticipated a 2016 fuel cost of

302 $1,078,088,287. The unaudited actual expenditures were $1.210 billion, a 20 percent increase.

303 It is with a high degree of certainty that we conclude that PREP A will not achieve the $1.581 to

304 $1.686 billion in savings from oil price declines from FY 2014 to FY 2017 identified in PREP A's

305 (seemingly inconsistent) presentation. If PREP A's fuel cost remains at or near the 2016 level,

306 PREP A's actual cost for 2017 will be as much as $400-$500 million more than the proposed budget

307 submitted in support of the Revenue Requirement.

45 Indeed, PREPA's August monthly report shows fuel costs in the first two months of FY 2017 have declined only 12% relative to the same period of FY 2016 (p.2). Additionally, this document shows that actual fuel costs in the first two months of FY 2017 were more than 100% above budget "due to higher cost of fuel than budgeted" (p. 8). PREP A Monthly Report to the Governing Board, August 2016. http://www.aeepr.com/INVESTORS/DOCS/Financial%201nformation/Monthly%20Reports/2016/August%202016.p

df 46 Energy Information Administration, Short-Term Energy Outlook (Crude Oil), October 2016 http://www. eia .gov /forecasts/steo/ 47 Knoema, "Crude oiJ· price forecast: Long term 2016 to 2025", https://knoema.com/yxptpab/crude-oil-priceforecast-long-term-2016-to-2025-data-and-charts 48 PREP A Supplemental Integrated Resource Plan, April19, 2016. Appendix C.

Page 18

308 Q. YOU REFERENCED REVENUE-FUNDED CAPITAL EXPENDITURES AS A THIRD

309 FACTOR THAT COULD DRIVE RATES HIGHER THAN PROJECTED. PLEASE EXPLAIN.

310 A. Another factor that is likely to drive rates higher than forecast is the likelihood that PREP A will

311 need to continue financing capital expenditures through revenues, rather than through new debt

312 issuances, because of its inability to access the capital market. Schedule F-2, PREP A's balance

313 sheet under the proposed new rates, shows "New Issue Capex Financing" beginning in FY 2018

314 at a level of approximately $400 million per year tln·ough FY 2020. Presumably, this has been

315 included in the proposed rates through debt service charges, although this has not been presented

316 in a transparent manner. However, if PREP A is not able to access the capital markets beginning in

317 FY 2018, PREP A would need to finance this level of capital investmenttln·ough cash, which would

318 require a higher revenue requirement in those years than debt financing, leading to upward pressme

319 on rates. (PREPA's IRP estimated the cost of AOGP at $385 million 49 and, given that the

320 Commission has ordered PREP A not to proceed with construction at this time, some of this

321 projected "new issue capex financing" could presumably be reduced).

322 In fact, it is unlikely that PREP A will be able to enter the capital markets starting in FY 2018. The

323 Commission noted in its IRP Order that "[i]t is unce1iain when PREP A will have access to the

324 capital markets; and when it does have that access, in what amounts and at what cost." (IRP Order

325 of September 23, 2016 at paragraph 65). The testimony of PREP A witnesses Pampush, Porter and

326 Stathos estimates that PREP A will be able to regain access to the capital markets at reasonable

327 rates "by 2020 or later" (lines 991-996). The testimony of these witnesses suggests that Schedule

328 F's representation of"new issue capex financing" beginning in FY 2018 is unlikely to materialize.

49 PREP A Integrated Resource Plan (August 17, 2015 version), page 5·7.

Page 19

329 Q. YOU MENTIONED THE TRANSITION CHARGE AS A FOURTH FACTOR PUTTING

330 UPWARD PRESSURE ON RATES. PLEASE EXPLAIN.

331 A. The costs embedded in the Transition Charge, including the upfront and ongoing financing

332 costs for the Securitization Bonds, are not subject to Commission oversight and are not being

333 reviewed in this proceeding. The amount of up front fees is already well over the budget initially

334 provided in PREP A's application for the Transition Charge. The Commission's statutory lack of

335 oversight over the fees, combined with appm-ent conflicts of interest in establishing the fees, has

336 led the Commission to be concerned that PREP A ratepayers will be exposed to "fees without

337 limit."50 An increase in the Transition Charge would result in higher overall rates.

338 Additionally, as noted by PREP A witness Donahue in response to the Commission's 4th request

339 for infmmation, there is a risk that the final structure of the bond deal could result in higher rates.

340 Specifically, if more bondholders pmiicipate in the underlying securitization transaction, then the

341 Transition Charge would have to rise to cover the additional bond indebtedness of the Corporation.

342 But, Donahue is "unclear" whether that transfer of indebtedness from PREP A to the Securitization

343 transaction would decrease PREP A's total debt service obligations.51 She states, "it is possible

344 that the mix of pmiicipating bonds will have a different maturity and interest profile than

345 previously assumed, which would malce the ultimate impact of greater participation [in the

346 Restmcturing Support Agreement] unclear. .. [W]e cannot assume a $1 for $1 decrease." In other

347 words, there is a risk that the Transition Chm·ge could be increased without an equal decrease in

50 Restructuring Order (June 21, 2016) paragraph 266. 51 PREP A Response to CEPR-SGH-02-02 (a).

Page 20

348 the legacy debt service component of the base rate, depending on the final participation level in

349 the Restructuring Support Agreement, thereby driving rates up. 52

350 III. PREP A's proposed expenditures and investments require close regulatory scrutiny.

351 Q. DO YOU AGREE THAT PREP A REQUIRES CLOSE COMMISSION OVERSIGHT OVER

352 ITS EXPENDITURES AND INVESTMENTS?

353 A. Yes. PREP A witness Hemphill states that PREP A's proposed formula rate-making (FRM)

354 mechanism will provide for "increased Commission oversight of PREP A's business plam1ing

355 process." (Hemphill supplemental testimony lines 53-54).

356 We agree that the many changes underway at PREP A, the apparent low level of trust between

357 PREP A and its regulator53, and our concerns described previously regarding likely slippage in

358 PREP A's budget initiatives all point towards a need for strong Commission oversight.

359 In this section, we evaluate PREPA's proposed FRM mechanism in that light and provide

360 additional recommendations for the Commission's consideration.

361 Q. WHAT IS THE PROPOSED "FORMULA RATE-MAKING" MECHANISM?

362 A. PREP A has proposed a "fmmula rate-making" (FRM) mechanism in which PREP A would file

363 a base rate case every three years, and in the intervening years it would true-up the different

364 components of the rate to align with actual costs without changing the allocation of costs between

365 customer classes. Specifically, in the years in which it does not file a base rate case, our

366 understanding is that PREP A would seek to set rates for the next fiscal year based on its budget

52 Ibid. 53 As suggested by the Commission's recent IRP Order (see paragraphs 13, 142, 243, 281) and Transition Charge Order (see paragraph 272).

Page 21

367 for that fiscal year, plus a reconciliation of the previous year's rates based on actual expenditures

368 and the Commission's determinations of prudence. Ifrates in the previous year were too low to

369 cover expenditures, and the Commission fmds that PREP A's expenditures were prudent, rates

370 would be raised in the subsequent year to cover the difference. This reconciliation would only

371 apply to the base rate, i.e. it would not include the fuel and purchased power adjustment riders, the

372 Transition Charge, CILT and subsidies which are to be adjusted separately. (Hemphill Direct

373 Testimony, Ex. 7 lines 404-418).

374 PREP A is proposing a six-month process for adjusting the rates in the off years (the years in which

375 it is not filing a base rate case). This is the same length of time that Act 57 allows for a base rate

376 case (as the current rate case proceeding), although the Commission has the option of a 60-day

377 extension for a base rate case. PREP A contemplates an annual proceeding that would allow for

378 intervention by other patties, discovery and testimony (Hemphill Supplemental Testimony, Ex. 16

379 lines 190-192).

380 Q. DO YOU AGREE WITH REFERRING TO THIS PROPOSAL AS A FORMULA RATE?

381 A. No. The concept of "formula tate-making" catTies the implication of less regulatory oversight

382 and even automatic adjustments to rates according to a formula. 54•55 We strongly oppose any such

54 Indeed, witness Hemphill's original direct testimony (lines 353-356) seemed to support this concept by describing the proposed FRM as "a cycle where rates are revised every year to reflect updated cost and usage information with an in-depth examination of the cost components, allocation studies, interclass revenue allocation adjustments and rate design occurring every three years" (emphasis added). This description of the FRM does not appear to contemplate providing in-depth information to support the costs that PREP A would seek to recover. However, the Hemphill supplemental testimony calls for full Commission oversight of the cost components of the revenue requirement (lines 185-195).

55 Indeed, one of the examples of formula rates cited in witness Hemphill's direct testimony provides an extreme example of limited regulatory oversight and public scrutiny. Alabama's FRM mechanism has been criticized for lack of transparency, cursory Commission review and excessive profits to Alabama Power (See: D. Schlissel and A.

Page 22

383 concept. At this time in which so many changes are being attempted within PREP A and a major

384 attempt is nndetway to reform the agency, the public cannot afford less regulatory oversight of

385 PREP A Additionally, automatic adjustments to rates would likely result in excessive rate

386 increases in the absence of Commission oversight over PREP A expenditures.

387 Witness Hemphill's supplemental testimony, however, states that there would be no difference in

388 the amonnt of information provided to the Commission regarding the components of the revenue

389 requirement to be adjusted under its proposed annual rate filings, and that PREP A is committed to

390 providing audited financial information in annual rate cases as available56 Additionally, PREP A

391 proposes that the level of review be similar, with a 180-day process and full intervenor

392 pmticipation. The difference is that a base rate case can be extended to 240 days at the

393 Commission's discretion, providing for a higher level of scrntiny and oversight. (We would

394 recommend, if the Commission adopts some version ofPREPA's proposal, that it allow the option

395 for a 60-day extension in an annual rate review case).

396 If a proposal in line with the recommendations in witness Hemphill's supplemental testimony that

397 does not involve less regulatory oversight, nor any automatic adjustment to rates, is considered, it

398 should more clearly be refened to as an "annual rate review", not a fmmula rate. We will refer to

399 it as an "annual rate review" in the rest of this section and urge the Commission to reject the

400 language of"formula rate-making."

Sommer, "Public utility regulation without the public: The Alabama Public Service Commission and Alabama Power,"

Institute for Energy Economics and Financial Analysis, March 1, 2013.) Clearly this outcome should be avoided. 56 Hemphill Supplemental at lines 333-335.

Page 23

401 Q. DO YOU AGREE THAT PREP A'S UNIQUE CIRCUMSTANCES CREATE CHALLENGES

402 FOR TRADITIONAL RATE-MAKING?

403 A. Yes. The logic of h·aditional rate regulation based on a historical test year will not work well

404 for PREP A under its cunent circumstances. Under this type of regulation, a utility does not begin

405 to recover on capital expenditures made after its last rate case until its next rate case. In an

406 environment of growing sales, and hence growing revenues, the utility is able to cany the cost of

407 these investments without the need to increase rates. But PREP A is not operating in an

408 environment of growing sales, nor does it have the ability to debt fmance its capital expenditures;

409 it must pay for them immediately out of revenues. 57

410 Q. WHAT ARE THE DIFFERENCES BETWEEN A BASE RATE CASE AND PREPA'S

411 PROPOSED ANNUAL RATE REVIEW MECHANISM?

412 A. Assuming that the Commission requires the same level of detail on the cost components of the

413 revenue requirement as it would in a base rate case, there are two main differences between a

414 traditional base rate case and the proposed annual rate review. One is that PREP A would not be

415 presenting a cost of service study or proposing changes to rate design when it files an mmual rate

416 review case. Instead, as required by law, PREP A will initiate proceedings to adjust the rate design

417 every tlu·ee yem·s with all requirements including cost of service studies.

57 PREP A can also leverage third-party financing for major capital projects, via power purchase agreements, for example. But PREPA still has significant ongoing capital expenditures that currently must be funded through revenues.

Page 24

418 The other difference is that under the annual rate review mechanism, PREP A will set rates based

419 on its budget for the next yeaT and based on reconciling the previous year's revenues with actual

420 costs. A traditional base rate case does not involve such a reconciliation.

421 Q. WOULD THE RECONCILIATION OF THE PRIOR YEAR'S REVENUES AND COSTS BE

422 AUTOMATIC?

423 A. Not as proposed in witness Hemphill's supplemental testimony, and we agree that the

424 reconciliation should not be automatic. The Commission would have the opportunity to determine

425 whether the incun·ed costs were prudent (Hemphill Ex. 16 at lines 98-104). If the Commission

426 dete1mines that PREP A incuned certain costs impmdently, they would be excluded from cost

427 recovery.

428 However, we note that - because PREP A does not have owners' equity - even if a cost is

429 disallowed as imprudent, ratepayers will still pay indirectly for that cost, through deferred

430 maintenance or investment. That is, ifPREPA's operational expenditures are higher than budgeted

431 in a given year, PREP A will have to adjust by reducing maintenance OT capital expenditures in that

432 year. If the Commission disallows PREP A's excessive operational expenditures as impmdent, then

433 PREP A will never recoveT the funds that it should have spent on maintenance or capital

434 expenditures. In other words, unlike a private investor-owned utility, a determination of

435 imprudence does not mean that shareholders bear the cost; it means that ratepayers bear the cost

436 in another form. Therefore, it is impmiant that PREP A be regulated as closely as possible to

437 minimize imprudent expenditures.

438 Q. DO YOU HAVE ANY RECOMMENDATIONS FOR HOW COMMISSION MIGHT

439 ATTEMPT TO MINIMIZE IMPRUDENT EXPENDITURES?

Page 25

440 A. Yes. The Commission could requne more fi·equent financial updates on maJor capital

441 expenditures, such as AOGP if construction IS ultimately approved. This would allow the

442 Commission to detect any problems with the project early and also to ensure that PREP A is

443 structuring contracts appropriately so that engineering, procurement and construction contractors

444 are bearing some of the risks of the project going over budget.

445 Additionally, the Commission could hire its own engineering advisor to oversee PREPA's

446 management of large projects, such as AOGP. This has occuned elsewhere; for example, the

447 Mississippi Public Service Commission has an engineering advisor monitoring the Kemper

448 integrated gasification combined cycle project. 58

449 The Commission conld also require PREP A to present a turnkey cost estimate from an engineering,

450 procurement and construction (EPC) contractor for major projects. In a turnkey project, the EPC

451 contractor is charged with delivering the final project at a set cost. Requiring a turnkey estimate

452 would give the Commission a reference point for how an independent third-party would price the

453 risk inherent in a large construction project and assist the Commission in dete1mining what is a

454 prudent cost.

58 "URS Corporation (URS), later acquired by AECOM, was requested by the Mississippi Public Service Commission (MPSC) to provide Independent Monitoring services for the Kemper Integrated Gasification Combined Cycle (IGCC) Project located in Kemper County, MS. The scope of services includes monthly reporting by URS (AECOM) and its subcontractors, the Independent Monitor (IM), of the status and prudency of the on-going engineering, procurement, construction and startup activities performed by Mississippi Power Company (MPC or the Company), its parent Southern Company and subsidiary Southern Company Services (SCS), and its subcontractors on the project." (URS Corporation, IM Monthly Report to Mississippi Public Service Commission, May 2016, http://www.psc.state.ms.us/executive/pdfs/2016/Kemper/Monthly%20Report%20May%202016%20Executive%20 Summary. pdf)

Page 26

455 Finally, the Connnission could investigate the oversight modellmown as the Independent Private

456 Sector Inspector General (IPSIG). 59•

60

457 These reconm1endations could be adopted whether or not an annual rate review is approved.

458 Q. COULD THE PROPOSED ANNUAL RATE REVIEW BE USED BY THE COMMISSION

459 TO FURTHER THE GOAL OF MINIMIZING IMPRUDENT EXPENDITURES?

460 A. Due to the serious inconsistencies and flaws in this petition, we believe that the Commission

461 should reject the proposed rate increase, which would preclude the Connnission from adopting an

462 annual rate review mechanism at this time. However, if and when the Connnission approves what

463 it considers to be a just and reasonable base rate for PREP A, strict oversight must follow and an

464 annual rate review could be used by the Commission as a tool for minimizing imprudent

465 expenditures going forward.

466 The proposed annual rate review has the advantage of setting a calendar for consistent and fairly

467 frequent filings from PREP A. In order to avoid imprudent capital expenditures, we urge the

468 Connnission to require a detailed budget of capital expenditures for the next fiscal year and explain

469 any deviations fi·om PREP A's approved IRP.

470 The Connnission can also minimize the risk of imprudent expenditures by using its authority to

4 71 disallow imprudently incuned costs. If PREP A believes that it will have the opportunity to raise

472 rates in the next year if it misses its budget targets, PREP A will have less incentive to hit those

473 targets. This risk can be mitigated if the Commission insists on a high level of transparency up-

59 http://www.iaipsig.org/directors.html 60 See: http://www.osc.state.ny.us/reports/nyra/nyrareport905.pdf for an example of how an IPSIG works. Tom Sanzil/o, the former First Deputy Comptroller of New York State, participated directly in this effort to reform a public authority that for decades was mismanaged and ultimately subjected to criminal sanctions.

Page 27

4 7 4 fiont regarding how PREP A plans to meet its savings targets in the coming year and also makes

475 use of its ability during the reconciliation process to disallow expenditures that were not prudently

476 incuiTed.

477 IV. The proposed rate design does not give customers, particularly industrial and

478 commercial customers, the flexibility to lower their own energy costs and expand the use of

4 79 renewable energy generation in Puerto Rico

480 Q. HAS THE COMMISSION ARTICULATED ANY GOALS REGARDING THE

481 TRANSFORMATION OF PUERTO RICO'S ELECTRICITY GENERATION MIX?

482 A. Yes. The recent IRP order describes the Commission's goal as "to replace old, costly plants

483 with lower-cost options: more efficient plants, renewable resources, energy efficiency, demand

484 response and distributed generation technologies- some of which empower consumers to manage

485 their own costs, all of which reduce environmental damage as well as customers' exposure to fuel

486 price volatility." (IRP Order, September 23,2016 at paragraph 30).

487 Q. DOES PREP A'S PROPOSED TARIFF IN THIS CASE FURTHER THE ABOVE GOAL?

488 A. No. The proposed tariff submitted by PREP A in this proceeding runs counter to this goal, by

489 failing to allow customers to source power from lower-cost options through a wheeling service

490 and by discouraging customers from investing in distributed renewable energy generation

491 technologies.

492 Q. PLEASE EXPLAIN HOW A WHEELING SERVICE WOULD ALLOW CUSTOMERS TO

493 SOURCE LOWER-COST POWER.

Page 28

494 A. PREP A's generation costs are very high. PREP A's Embedded Cost of Service Study found that

495 a 100% cost-based unbundling of Tariff GRS, for example, resulted in the generation portion of

496 the rate being 11.5 cents/kWh (Zarumba and Granovsky Exhibit 4.0 at line 491 ). 61 This is a result

497 ofPREPA's expensive oil-based generation system (see Response to Commission Request CEPR-

498 PC-01-13). The few operational renewable energy contracts that PREP A has are also over-priced,

499 at an average price of$189/MWh for solar and $157/MWh for wind. This is a sharp contrast with

500 jurisdictions in the United States, even those with far less solar potential than Puerto Rico. In

501 Minnesota, for example, a state with a demonstrably lower solar resource than Puerto Rico, Xcel

502 Energy has estimated the cost of solar in 2016 at $67.30/MWh, nearly two-thirds less expensive

503 than PREP A's current solar contracts. 62 PREP A did not competitively bid these renewable energy

504 contracts, resulting in unnecessarily high prices.63 This strongly suggests that third-patty power

505 providers could provide renewable energy to commercial and industrial customers less expensively

506 than PREP A, if third-patty power providers had access to the grid.

507 If wheeling were allowed, it would also provide a strong incentive for PREP A to reduce its costs

508 and greater leverage to PREP A in renegotiating some of its above-market contracts64 because of

509 the introduction of competition. In the near term, the introduction of competition would allow

510 customers greater freedom to source electricity from renewable energy or other providers

511 potentially at a lower cost.

61 This is not the actual rate proposed for Schedule GRS because of mitigation and the inclusion of a $8/month fixed charge. 62 Xcel 2015 IRP Supplement filed with the Minnesota Public Utilities Commission, January 29, 2016. 63 A. Skibell, "How a stubborn utility and aging grid added to island's woes," E&E Publishing, May 2, 2016. 64 Though PREP A has signed a large number of above-market contracts for solar projects, it is unclear how many of these projects will ultimately be developed. (IRP Order, September 23, 2016, paragraphs 184-188}. PREP A has been ordered to renegotiate or terminate above-market contracts for projects that are not operational. (IRP Order, September 23, 2016, paragraph 299}.

Page 29

512 In addition, wheeling (provider selection) at the distribution system level for renewable energy

513 would increase demand from renewable sources, providing for significant opportunities in the

514 application and development of technologies such as utility scale batteries, and additional storage

515 and smmi grid technology placing Puerto Rico as the natural leader in its region.

516 Additionally, we note that this is an opportune time for the introduction of a wheeling tariff because

517 PREP A is in the early stages of major capital investment in modernizing its generation system, but

518 it has not yet made significant investments. The Modified IRP ordered by the Commission

519 provides for a more flexible approach than PREP A originally proposed by ordering the near-term

520 construction of smaller units at Palo Seco and increased energy efficiency and demand response.

521 With proper planning, PREPA could anticipate the depmiure of commercial and industrial

522 customers from its system under a wheeling tm·iff (to purchase electricity from third-pmiy

523 renewable energy or highly efficient fossil generation sources) and not overbuild its own

524 generation system.

525 Q. HAS PREP A MADE ANY STEPS TOWARDS THE ESTABLISHMENT OF WHEELING

526 SERVICE?

527 A. The unbundling of tariffs proposed in this proceeding is one step towards the establishment of

528 a wheeling service. The testimony of witnesses Zarumba and Granovsky acknowledges the

529 importance of tariff unbundling because "different customers purchase different services from the

530 utility" (line 439). Zarumba and Granovsky at lines 467-482 provide two examples of tariff

531 unbundling in the mainland United States: the Federal Energy Regulatoty Conm1ission's

532 requirement since 1996 of"open access to transmission" and the decisions of seventeen U.S. states

533 to allow retail choice. Both of these examples were designed to further competition between

Page 30

534 generation sources by allowing third-party power providers fair and open access to the grid and

535 allowing customers to purchase directly from these providers, bypassing the generation owned by

536 their incumbent utility. Despite providing these examples of tariff unbundling, PREP A is still not

537 proposing to allow customers to choose the services (generation, transmission and distribution)

538 tbat they can purchase fi·om PREP A. PREP A does not offer, and does not plan to offer,

539 transmission and distribution-only services that would give open access to third-party power

540 providers and permit customers to contract with those providers.

541

542 Q. IS THERE ANY LEGISLATION THAT REQUIRES WHEELING?

543 A. Yes. Our understanding is that Puerto Rico Act 73-2008 directed PREP A to establish a wheeling

544 service by January 2, 2010. This did not occur. Later, through Act 57-2014 (Section 6.30), the

545 Connnission was required to regulate wheeling and to establish tbe rules necessary for a wheeling

546 service.

547 PREP A should be required in this proceeding to develop a wheeling tariff so that commercial,

548 industrial, and if possible residential customers can take advantage of renewable energy sources

549 that are lower cost tban those supplied by PREP A. This would assist customers in controlling costs,

550 as well as spuning the development of renewable energy in Puerto Rico.

551

552 Q. PLEASE EXPLAIN WHY YOU BELIEVE THAT THE PROPOSED TARIFF

553 DISCOURAGES INDUSTRIAL CUSTOMERS FROM INVESTING IN DISTRIBUTED

554 RENEW ABLE ENERGY TECHNOLOGIES.

555 A. PREP A's proposed tariff also disincentivizes industrial and commercial customers from

556 investing in distributed renewable energy through its over-reliance on demand charges and non-

Page 31

55 7 bypassable energy charges. The proposed rate increases the prop01iion of revenues fi·om demand

558 chaTges versus energy charges. While the overall industrial rate increase is 26.2%, the increase in

559 demand charges for the main industrial taTiff, Schedule GSP (fi·om $8.1/kVA to $12/kVA) is

560 neaTly 50%. The net metering credit that industrial and commercial customers receive fi·om

561 investing in their own distributed renewable energy generation is equal to the energy-only charge

562 (11.1 cents/kWh for GSP Tariff), not the retail rate. In other words, net metering customers still

563 must pay the demand charges, as well as the CIL T, subsidy chaTge and the securitization charge.

564 In tariff GSP, for example, these charges amount to $12/kVA (demand), $200/month (fixed

565 chaTge), 0.303 cents/kWh (CILT), 1.02 cents/kWh (subsidy chaTge) and 3.05 cents/kWh

566 (Transition ChaTge ), none of which are subject to the net metering credit. This taTiff design, which

567 fails to give any capacity credit to distributed renewable energy resources, fails to recognize the

568 benefits that distributed renewable energy customers provide, including avoided line losses,

569 deferred transmission and distribution capacity upgrades, deferred generation capacity, and

570 reduction in peak demand. 65

571 The tariff design also highlights the unsustainably high debt levels embedded in PREP A's rates.

572 Overall rates for GSP customers are proposed to be approximately 20.4 cents/kWh. 66 A net

573 metering customer on this taTiff can only avoid 54% of that charge, 11.1 cents/kWh. An industrial

574 or commercial customer on schedule GSP that offsets all of its own electricity consumption will

575 still pay the equivalent of9.3 cents/kWh, of which 4.4 cents/kWh aTe subsidies and the legacy debt

576 embedded in the securitization charge.

65 Because of PREP A's load shape, which includes an afternoon peak and an evening peak, distributed solar resources would need to be combined with an effective demand response program in order to shift load away from the evening

peak so that solar can contribute to reducing peak demand. 66 Total revenues for class GSP ($920 million) divided by total sales (4,510 GWh), as provided in Schedule H.

Page 32

577 Effectively, by weakening the incentive for industrial and commercial customers to invest in their

578 own distributed renewable energy resources and instead tying them to payments of PREP A's

579 legacy debt, PREP A's proposed tariff is crowding out investment in renewable energy generation

580 in Puerto Rico.

581 V. PREP A's assertion that proposed rates will maintain an adequate debt service coverage

582 ratio is flawed.

583 Q. WHY IS THE DEBT SERVICE COVERAGE RATIO (DSCR) THE CRITICAL CREDIT

584 METRIC FOR PREP A'S RE-ENTRY INTO THE CREDIT MARICETS?

585 A. We concur with PREP A on three critical points. First, that the DSCR is the most important

586 credit metric used by credit rating agencies to establish the Authority's creditworthiness at this

587 time67 Second, that Modified Cash Basis accounting is an approach 68 that can be utilized for a

588 public power entity with inadequate cash flow and constrained market access. We take note of

589 PREP A's use of this method to ensure that PREP A "recovers prior year capital expenditures

590 through debt service and anticipated capital expenditures through revenue funded capex."69 The

591 Modified Cash Basis method in theory allows the Authority to capture and present its full revenues

592 and expenses in a transparent manner. Third, that "the true test of credit worthiness is the belief by

593 lenders that they will receive timely and complete repayments of all of the cash flow that are due

594 them. This analysis cannot capture subjective beliefs, but can provide values relative to

595 quantitative indicators. "70

67 Pam push, Porter and Stathos Testimony, Line 1359-1360. 68 Pampush, Porter and Stathos Testimony, Lines 296·298 69 Pam push, Porter and Stathos Testimony, line 413. 70 Pam push, Porter and Stathos Testimony, Lines 1166·1169.

Page 33

596 Q. HOW IS THE DSCR DEFINED?

597 A. The DSCR establishes the relationship between PREP A's Net Income (available resources to

598 pay debt and invest after revenues are subtracted from expenses) and its Total Debt Service

599 obligation. The relationship is described as a ratio and it is derived from the typical formula:

600 Debt Service Coverage (DSCR) =Net Income/Total Debt Service

601 Net Income is measured as revenues minus operating expenses. When PREPA's Net Income

602 exceeds Total Debt Service the ratio is greater than I and demonstrates that PREP A has sufficient

603 cash to pay its debt service.

604 Q. WHAT IS CONSIDERED AN ADEQUATE DSCR?

605 Credit rating agencies, lenders and PREP A consider a 1. 0 DSCR an inadequate level of coverage

606 to determine credit wmihiness. Utilities or any business face risks that alter their budgets and

607 financial plans in unforeseen ways. The investment world looks to higher ratios, showing more

608 evidence of cash availability, to offset these risks. PREP A's Trust agreements require 1.2 DSCR. 71

609 Given ·PREP A's currently distressed fiscal condition and its lack of reasonable market access

610 PREP A's consultants have concluded that the Authority would fare better in the capital markets

611 with a DSCR of !.57 to 2.00.72

71 Pam push, Porter and Stathos Testimony, Lines 1376-1379. 72 Pam push, Porter and Stathos Testimony, Lines 1038

Page 34

612 While PREPA's financial presentations offer a number of different DSCR valuations in

613 testimony73 PREP A infmmed the Commission that the DSCR is 3.974 for its stand-alone Legacy

614 Debt. Then, when PREP A's Legacy Debt is consolidated with the Securitization Charge, PREP A

615 claimed that the DSCR is 1.9. The underlying spreadsheet was later corrected and PREP A revised

616 its estimate of the DSCR for consolidated debt to 1.47.75

617 PREP A contends that the post rate increase financial scenario suppmis a finding that the rate

618 increase(s), individually and when consolidated, supply PREP A with sufficient revenue to pay

619 debt service. In short PREP A finds that its finances exceed the 1.57 to 2.00 DSCR and meets the

620 .standard most conducive to the Authority's re-entry into the capital markets.

621 Q. DO YOU CONCUR WITH THIS CONCLUSION?

622 A. No. Our analysis supports the following three findings:

623 1. The Authority does not meet the DSCR standard when presented on a consolidated basis.

624 PREP A's own analysis shows a DSCR of 1.47. This falls below its own standard of 1.57.

625 2. Our analysis shows the DSCR for consolidated debt is 1.36 (not 1.47 as PREP A claims). If

626 capital expenditures (which are costs that are passed to consumers) are included in the DSCR

627 calculation, as they should be, the DSCR is even lower.

73 Pam push, Porter and Stathos Testimony, Lines 802-804. 74 This portion of the testimony is supported by a PREPA spreadsheet: "PREPA Rate Case Financial Model

160620_Rate Change to PR", Tab: DSCR, cell H15. "PREP A RCFM". 75 Pam push, Porter and Stathos Testimony Line 803 refers to DSCR of 1.9 for PREP A's securitized debt inclusive of its legacy debt. The spreadsheet supporting this figure is: "PREP A Rate Case Financial Modei160620_Rate Change to PR", Tab: IS, HS9, "PREP A RCFM". This number was later corrected by PREP A: see and Pam push, Porter, Stathos, Supplemental Direct Testimony, Exhibit 14.00, October 13, 2016, Lines 73-83 and PREP A Exhibit 14.02, Tab IS, HS9.

The corrected DSCR is 1.47.

Page 35

628 3. Taken as a standalone entity PREP A (non-securitized legacy debt) only meets the standard if

629 the Commission:

630 a. accepts unrealistic and risky budget forecasts; and

631 b. excludes capital expenditures that are being passed along to consumers in the final

632 rate design.

633 Using more realistic budget assumptions the DSCR is below the 1.57 standard.

634 Q. HOW DID YOU REACH THIS CONCLUSION?

635 A. IEEFA has constructed a post rate increase DSCR based upon PREPA's presentation of

636 revenue, expenses and its debt service needs in Schedule A (REV). We present two DSCR

637 coverage ratios using the same methodologies employed by PREP A in the PREP A RCFM

638 spreadsheet. The first covers a consolidated revenue and expense scenario that includes the

639 revenue and expenses for both PREP A under its rate request in this proceeding and the

640 Revitalization Corporation as presented and approved by the Commission in the Securitization rate

641 docket (CEPR-AP-2016-001). The second scenario excludes the Transition Charge and assesses

642 PREP A's DSCR as a standalone financial entity.

643 Q. HOW DOES PREP A'S DEBT SERVICE COVERAGE RATIO FOR ITS CONSOLIDATED

644 DEBT SCENARIO COMPARE TO ITS STANDARD OF 1.57?

645 A. PREP A's standard which would provide the Authority with an improved chance of market re-

646 entry is a DSCR of 1.57. PREP A's revised analysis contained in Exhibit 14.02 sets the DSCR at

647 1.4 7, below the credit metric necessary to support the Authority's market re-entry.

Page 36

648 Q. PLEASE EXPLAIN HOW YOU CALCULATED A DSCR OF 1.36 (AS OPPOSED TO

649 PREPA'S ESTIMATE OF 1.47) WHEN PREPA AND THE CORPORATION ARE

650 CONSIDERED AS A CONSOLIDATED ENTITY.

651 A. Our analysis and conclusions start with PREPA's budgetary submissions that suppmt a

652 consolidated Revenue Requirement of $3,462,194,772 against projected operational expenses

653 (minus revenue funded capex and Legacy Debt Service) of $2,346,907,833 as provided in

654 Schedule A-1 (REV). IEEFA then subtracts revenues fi·om expenses to derive a Net Income of

655 $1,115,286,833 (See Table 3).

656 Table 3: Post Rate Increase Consolidated 2017 Revenue and Expenses 76 and IEEFA 657 Calculation of Net Operating Income

Budget Item Total

Revenues $3,462,194,772 Expenses $2,346,907,833

IEEFA Net $1,115,286,889 Income

658 Table 4 uses the Net Income from Table 3 and adds Total Debt Servrce by combining the Debt

659 Service values provided in Schedule A-1 (Rev) for PREP A's Legacy Debt Service and Transition

660 Charge. The consolidated DSCR is 1.36. Tllis is below the standard established by PREP A (1.57

661 to 2.00).

662 Table 4: Post Rate Increase IEEFA Calculation of 2017 Debt Service Coverage for 663 Consolidated PREP A Legacy Debt and the Securitization Charge

Debt Service Net Operating Total Debt Service Debt Service Income Coverage Ratio

(DSCR)

76 The revenue and expenses are taken from Exhibit 5.04 and Schedule A-1. The Net Income calculation is derived by lEE FA as a simple subtraction of revenue and expenses.

Page 37

Consolidated PREP A $1 '115,286,889 $817,653,975.00 1.36 and Secudtization)

664

665 It is important to note that this calculation is very conservative because it does not include capital

666 expenditures at all. In reality, 2017 revenue funded capital expenditures should be included in this

667 calculation as they represent a 2017 cash cost that PREP A must pay. Either capital expenditures

668 should be treated as an operating expense (because they will be paid in cash through revenues in

669 FY 2017), which would reduce PREP A's net operating income to $778,729,081 and the DSCR to

670 0.95; or, the expenses should be included in debt service as they are integral to PREP A's overall

671 debt management plan.

672 Q. CAN YOU EXPLAIN HOW PREPA CALCULATED A HIGHER DSCR FOR THE

673 CONSOLIDATED ENTITY?

674 A. Yes. PREP A's Exhibit 14.02 shows PREP A revenues for FY 2017 of $2,997,855,381. When

675 added to the Transition ChaTge revenue requirement of $503,264,236, the total revenue for the

676 consolidated entity is $3,501,119,617. This is higher than the total revenue requirement shown in

677 Schedule A-1 (REV) for FY 2017, which we used in our calculation. We are unable to account for

67 8 the discrepancy.

679 Q. WHAT DO YOU CONCLUDE ABOUT PREPA'S ABILITY TO MEET ITS DSCR

680 STANDARD FOR ITS LEGACY DEBT WHEN THE AUTHORITY IS CONSIDERED AS A

681 STAND-ALONE ENTITY?

682 A. PREP A does not meet the standard uuless unacceptable and unrealistic budgetary assumptions

683 contained in the Authority's financial presentation are accepted.

Page 38

684

685

686

687

688

689

690

691

692

693

694

1695

696

697

698

699

700

701

702

We find the DSCR for PREP A's Legacy Debt is 1.95, but only if a number of speculative

assumptions regarding PREP A's budget are accepted, and if capital expenditures are not included

in the DSCR calculation. Without maldng these assumptions, the DSCR is less than 0.88.

Q. PLEASE EXPLAIN HOW PREP A'S ASSUMPTIONS LEAD TO A DSCR OF 1.95 WHEN

PREPA IS CONSIDERED AS A STAND-ALONE ENTITY, RATHER THAN PREPA'S

ESTIMATE OF 3.9.

A. PREP A's 3.9 DSCR represents the Authority's Legacy Bonds only77 This analysis and this

proceeding is concerned with PREP A's total legacy debt service costs (exclusive of Transition

Charge revenues).

In order to maintain a consistent method of presenting the DSCR for both the stand-alone and

consolidated calculations IEEF A stmts with the smne figures for revenue and expense contained

in -Schedule A-1 (REV). We make one adjustment to Table 3. We reduce the projected $3.462

billion revenue figure by the value of the Transition Charge: $503 million. The Transition Chm·ge

is the mnount PREP A pays to service the Securitization bonds. Under the statute and presentation

to the Commission PREP A simply passes through the revenues for debt service for the Transition

Charge to a payment agent for the bondholders. Therefore, this revenue is not available to PREP A

to pay for its operating expenses, legacy debt or capex. This calculation reduces the revenue left

to pay PREP A's operational needs to $2.958 billion. The net income for PREP A as a stand-alone

entity is $612 million (Table 5).

77 "PREP A Rate Case Financial Model 160620_Rate Change to PR", Tab: DSCR, cell HlS. "PREP A RCFM". Cell HlS presents the debt service to cover only PREPA's remaining portion of the legacy debt. Cell H16 presents the DSCR assuming the full debt service cost for PREP A that is inclusive of the legacy debt and PREPA's other debt obligations. This DSCR is 2.07. The use of the 3.9 DSCR is somewhat misleading in the testimony as it does not present the impact of the full revenue requirement on the DSCR.

Page 39

703 Table 5: Post Rate Increase IEEFA Calculation 2017 Revenue and Expenses and Net 704 Operating Income for PREl' A on a Stand-Alone Basis

705

706

1707

708

709

710

711

712

Budget Item Total IEEFA Revenues $2,958,930.536 Expenses $2,346,907,833

IEEFA Net $612,022,703 Income

We then use the same DSCR equation employed in Table 4 to derive the debt service ratio for

PREP A as a stand=alone entity. The Legacy Debt requires an outlay of $314 million, resulting in

a debt service ratio of 1.95. This DSCR meets the standard of 1.57-2.0 established by PREP A, but

is lower than the 3.9 ratio presented in the Pampush, Porter and Stathos testimony (Exhibit 5.0).

713 Table 6: Post Rate Increase IEEFA 2017 Debt Service Coverage by for PREP A as standalone

714 entity

715

Debt Service Net Operating Total Debt Se1-vice Debt Service Income Coverage Ratio

(DSCR) Legacy Funded $612,022,703 $314,389,739.00 1.95 Debt Service (LDS)

716

717 Q. PREP A'S EXHIBIT 14.02 (TAB IS, CELL:H48) DERIVES A 2.07 DSCR FOR PREP A'S

718 STANDALONE DEBT POSITION. DO YOU AGREE?

Page 40

719 A No. We use the same methodology as found in the spreadsheet, however we are using the data

720 presented in Schedule A-1 (REV).

721 Q. DO YOU BELIEVE THERE SHOULD BE ADWSTMENTS TO THE 1.95 DSCR FOR

722 PREPA'S REVENUE FUNDED LEGACY DEBT SERVICE FOR AN ACCURATE AND

723 PRUDENT PRESENTATION?

724 A Yes. As above, this estimate neglects capital expenditures, which do not appear to be included

725 in PREP A's accounting but should be incorporated in some fashion because they are a cost passed

726 to ratepayers. If PREP A's FY 2017 capital expenditures are included in operating expenses

727 (because they are to be funded through revenues), the net income on a stand-alone basis falls to

728 $275,464,895, implying a DSCR of only 0.88.

729 Additionally, the presentation contains assumptions related to oil pnces, PREP A's savmgs

730 initiative and energy sales that are speculative, even highly speculative. As discussed in Section

731 II of our testimony, we believe that PREP A has over-stated its savings initiatives, under-estimated

732 FY 2017 fuel costs and over-estimated future sales. Referring to Table 6, if PREP A's net income

733 drops by $120 million or more it will fail to meet the DSCR standard of 1.57. This is a very small

734 margin of error, given that we believe that PREP A's fuel costs, for example, may have been

735 underestimated by as much as $400-$500 million. Given the individual and cumulative impact of

736 the risks we cite above in Section II we conclude the actual DSCR for the legacy stand alone debt

737 is well below the 1.57 DSCR prescribed by PREP A's consultants.

738 Q. PLEASE SUMMARIZE YOUR DSCR CALCULATIONS

739 A The following table summarizes our DSCR calculations:

Page 41

740 Table 7: Summary of DSCR calculations

PREP A and Corporation PREP A stand-alone DSCR consolidated DSCR

PREP A estimate 1.47 3.9

IEEFA adjustments using 1.36 1.95; but < 1.57 if budget PREP A methodology and estimates not achieved assumptions IEEFA estimate accounting 0.95 0.88 for capital expenditures

741

742 Q. HOW SHOULD PREPA TREAT REVENUE-FUNDED CAPITAL EXPENDITURES IN

743 THE DEBT SERVICE COVERAGE CALCULATION?

744 A. We take no position on how the Commission should include the Capex Expenditures of $337

7 45 million for FY 2017 identified in the rate case. We only believe that it should be accounted for

7 46 somewhere in the calculation. If not, the presentation overstates the revenues available to pay debt

747 servrce.

7 48 All revenues and expenses need to be included in the calculation to provide a full and complete

749 DSCR. When the Revenue Funded Capex expenditure is properly included in some fonn into

750 PREP A's calculations, either as debt service or operating costs, it drives the DSCR down below

751 acceptable standards. In short, if the DSCR captures all revenues and expenditures then it will not

752 meet the standards. Or, in other words, PREP A only meets the DSCR standards if unrealistic

753 revenue, expenditure and energy sales assumptions are accepted, and if the Commission accepts

7 54 an incomplete and distorted financial accounting of the DSCR calculation that does not include

755 capital expenditures.

Page 42

756 Q. WHAT IMPLICATIONS DOES THE SUBSTANDARD DSCR HAVE FOR PREPA'S

757 BUDGET?

758 A. PREP A's low DSCR and umealistic budget forecasts mean that PREP A will face pressure to

759 raise its rates and lower its capital expenditures.

760 Under PREP A's plan, the first $503 million of electricity sales revenues go to pay the Transition

761 Charge to cover the bond indebtedness included in the Securitization transaction. As noted above

762 this arrangement substantially reduces PREP A's net income.

763 If PREP A's budget assumptions do not materialize, PREP A will need to figure out how to adjust

764 its budget and fmancial plan. If revenues decline due to decreased electricity sales, for example,

765 both the Transition Charge and the PREP A rate are placed under pressure. The choice for PREP A

766 management is either to increase rates or to reduce its budget in order to effectively supply

767 additional cash for the Transition Charge upward adjustments. If PREP A does not raise rates, it

768 will have shortfalls in its operational budget.

769 Similarly, if PREP A's expenditures exceed budget projections, the financial burden will fall upon

770 PREP A's operational budget and on the rates charged to customers.

771 As PREP A's consultants have noted, not only will this situation result in upward pressure on rates,

772 but it will also pressure PREP A to cut capital expenditures:

773 First, these (govemment-owned) utilities do not have owners' equity. Thus they are 77 4 considerably more sensitive to the fluctuations that are business as usual for any utility or 775 business for that matter. A swing in expense outside its control can wrealc havoc on the 776 utility's business plan. For PREP A, this means real delays in rebuilding and implementing 777 investment that ultimately makes them a more efficient utility.78

78 Hemphill Testimony, Lines 338-343.

Page 43

778 We concur that PREP A will likely be pressured to reduce its revenue funded capex initiatives in

779 2017 and beyond, as well as being pressured to increase rates to bring revenues in line with

780 expenses. This will slow down necessary projects for PREP A. We are also cognizant of the

781 negative impact that considerable budget slippage will have during 2017 on PREP A's ability to

782 re-enter the capital markets and implement future investment plans.

783 Q. ARE THERE OTHER AREAS OF CONCERN REGARDING PREP A'S ABILITY TORE-

784 ENTER THE CAPITAL MARKETS?

785 A. Yes. As described earlier in this section, we believe that PREP A's proposed rates will not result

786 in an adequate DSCR, leading to pressure to increase rates and cut capital expenditures.

787 However, PREPA's rates are already excessive, a factor that is not reflected in PREPA's

788 discussion of its ability to re-enter the capital markets. In July 2015 Moody's published a document

789 that addressed frequently asked questions regarding Puerto Rico's credit status. The document

790 provides a context for its credit rating of Puerto Rico given its extraordinary financial condition.

791 In addition to standard credit metrics, like the DSCR, the overall credit rating for Puerto Rico

792 must consider: bondholder recovery rates, missed bond payments, economic growth and revenues,

793 pension risk, federal response, sovereignty issues and bond insurance. The Moody's document

794 concludes that Puerto Rico's "main hurdle will likely be the ability to show its economy has

795 transcended the structural impediments to growth that have largely weighed on it since 2006."79

79 Moody's Investor Service, Frequently Asked Questions About Puerto Rico's Fiscal and Debt Crisis, Issuer-In-Depth, July 22, 2015, p.7

Page 44

796 The lack of any underlying economic data (see: PREP A's response to ICSE-PR question 5) to

797 support PREP A's implicit claim that the Puerto Rican economy can support the level of rates being

798 proposed is a broader impairment to PREP A's creditwmihiness.

799 Q. WHAT IS YOUR OVERALL CONCLUSION REGARDING PREP A'S PROPOSED RATE

800 INCREASE?

801 A. In this proceeding, the Commission is challenged to make a decision based on unreliable data

802 and flawed debt service calculation methodologies. The proposed rate increase is not affordable

803 and, even if it is implemented, PREP A is unlikely to realize the savings assumptions embedded in

804 the rate, increasing pressure to drive rates higher. We believe it will not be possible to achieve an

805 affordable rate structure without renegotiation of the underlying debt. Additionally the proposed

806 rate design undermines the goal of stimulating the development of renewable energy in Puerto

807 Rico.

808 For these reasons, we urge the Commission to reject the proposed rate increase and require PREP A

809 to go back to the drawing board and present a transparent and sustainable rate design. Moving

810 forward, we urge the Commission to closely monitor PREPA's expenditures and capital

811 investments, possibly in conjunction with the PRO MESA Board, to minimize the risk of imprudent

812 expenditures.

813 Q. DOES THIS CONCLUDE YOUR TESTIMONY?

814 A. Yes.

Page 45

iJ>tate ··~ ~· ....

CERTIFICATE OF AUTHENTICITY ~ ffdak c.~ s~~~ SZde~lde SZde~lm'~iY~ ~~dd

Rafael M Barker, State of West Virginia, serving in the capacity of Notary Public o:llin/for the State of West Virginia beginning Il/6/2009 and ending ll/20/2019, and that his acts are entitled to full faith and credit by the courts and authorities of the land, as contained in the records of my office.

No. C201610212543977

Secretary of State Bldg. 1, Suite 157-K 1900 Kanawha Blvd. East Charleston, WV 25305-0770

~Mt.ut ~ .4llrf' Aan/ ~ _;d, p/tea?' S.=/A'I _;d, $;tm ~ tf/.a~ tr~

Signature:

West Virginia Secretary of State

This Certificate only certifies the authenticity of the signature and the capacity of the person who has signed the public document, and, where appropriate, the identity of the seal or stamp which the public document bears. The Certificate does not certifY the content of the document for which it was issued. Thii Certificate is not valid for use anywhere within the United States of America, its territories or possessions.

Document Code: IPT Phone: 304-558-6000

886-767-8683 Visit us online or validate this document:

www.wvsos.com

A 'f'fES'f ATION

Affiant, Catherine Kunkel, being first duly sworn, states the following:

The foregoing Testimony constitutes the direct testimony of Affiant. Affiant states that she would give the answers set forth in the Testimony if asked the questions propounded therein at the time of the filing. Affiant further states that, to the best of her lmowledge, the statements made are true and correct.

Affidavit No. ___ _

Catherine Kunkel Consultant, ICSE-PR

Date: -----.:.1.:..cO/c:...2-"[/i'--'l"'-6 _

Aclmowledged and subscribed before me by Catherine Kunkel, of the personal circumstances above mentioned, in her capacity as Consultant for the ICSE-PR., who is personally !mown to me or whom I have identified by means of her driver's license number t11Li ~gt..j , in'

fj Charleston, West Virginia this c)}j!!! day of October 2016.

OFFICIAL SEAL STATE OF WEST VIRGINIA

NOTARY PUBLIC RAFAEl M. BARKER STATE !""ARM INSURANCE ·101 P~NNSYL.VANIA AVE. CHARLESTON, WV 2$302

My commflll.llon oxpitt~~ NIW~Iiibilt 21\ 20~9

Public Notary

'

STATE OF NEW YORK Ulster County Clerk's Office } SS.:

I, NinaPostupack, Clerk of the County of Ulster and of the Courts of said County, and ofthe Supreme Court ofthe State of New York, in and for said County, the same being Courts of Record, DO

. Avltu~ M. 'Sdw.ndc HEREBY CERTIFY. That .................................................................................................................. .

whose name is subscribed to the certificate of the proof or acknowledgement of the annexed instrument, and thereon, written, was at the time of taking such proof or acknowledgement, a Notary Public in and for the County of Ulster, dwelling in the said County, sworn and. duly authorized to take the same. AND FURTHER: That I am acquainted with the handwriting of such Notary Public, and verily believe that the signature to the said Certificate of Proof or acknowledgement is g-enuine.

IN WITNESS ~OF, I have hereunto set my hand, and affixed the seal of said Courts and County, this (1../ sf- day of ow h.e.-v , 20 I (a

Nina Postupack, County Clerk ---'-'~""""lU-"'t,..._· ~'-"~==.:...::.._· ~~Deputy Clerk

ATTESTATION

Affiant, Thomas Sanzillo, being first duly sworn, states the following:

The prepared Testimony I am sponsoring constitute the direct testimony of Affiant in

the above-styled case. Affiant states that I would give the answers set forth in the

Testimony if asked the questions propounded therein at the time of the filing. Affiant

further states that, to the best of my knowledge, the statements made are true and

correct.

Affidavit No. __ _

Acknowledged and subscribed before me byliit~/}ft(....<;,J/113!/ /Qofthe

personal circumstances above mentioned, in his capacity as Consultant for the ICSE-PR.,

who is personally known to me or whom I have identified by means of his driver's

/·license number?~ 6J'(.: fS, in 4 this 61-,{) day of f)cz.f 2016.

( >~~ \~-~ )( >&j0,_.~(x__ Public Notary AF\LEEN M. SCHENCK

Notary Public, State ol N?~ York F\eg. #01SC479990~ •

Qualilied in Ulster Gout •IY ;;JcJ 1 7 Commission Expires septern!Jer :00,

1 TESTIMONY OF VICTOR GLASS

2 Q. What is the name and position of the witness?

3 A. Victor Glass, Professor of Practice, Rutgers Business School

4 Director of Rutgers Center for Research in Regulated Industries (CRRJ)

5 CRRJ Scholar

6 Q. What is the witnesses' experience?

7 Victor Glass has more than thirty years of experience setting wholesale rates for more than

8 1100, mostly rural, telephone companies in the United States. He was responsible for annual filings

9 submitted to the Federal Communications Commission. He was also responsible for developing

10 rate designs that would accommodate the needs of companies with widely differing cost structures.

11 In addition, he worked with on developing new service tariff offerings. Besides rate-setting, Glass

12 was heavily involved in many proceedings that involved significant changes in regulatory design

13 in the telecommunications industry such as access restructure, rate of retum represcription and

14 universal service reform.

15 His academic ruticles have appeared in such joumals as American Economist, Energy

16 Economics, Government Infmmation Quarterly, Infmmation Economics and Policy, Joumal of

17 Business and Economic Studies, J oumal of Business Ethics Education, Journal of Business

18 Forecasting, Joumal of Regulatory Economics, and Rutgers Business Review. He has also been a

19 contributing author to several trade jownals including America's Network, OPASTCO Advocate,

20 Xchange, and Legg Mason's Industry Analysis Repmt.

21 Glass is cunently the director of the Center for Research in Regulated Industries (CRRl).

22 The Center was founded in 1977, located at Rutgers University. It aims to fmther the study of

23 regulation by reseru·chers in economics and finance, in academic and non-academic institutions.

1 1Page

24 The objective of the Center is to promote research, scholarship and education in regulatory

25 economics that have implications for public policy. Research and scholarship is promoted through

26 conferences, seminms, workshops, courses, publications and research projects. The Center has

27 been funded by registration fees, publication fees, sponsorship and research grants. Research in

28 electricity, gas, water, telecommunications, other network industries, and postal service has been

29 undertaken and promoted by the Center. Examples of the types of topics within the scope of the

30 Center are: regulatory governance mechanisms, pricing stmctures, etc. The Center has been

31 responsible for the publication of numerous books and the Journal of Regulatory Economics.

32 CRRI has organized workshops and conferences in the United States and Emope. Please refer to

33 http://www.business.rutgers.edu/crri for more infonnation.

34 Glass earned his PhD from Columbia University with specialties in econometrics and

35 monetmy theory. He also earned an MBA in finance and marketing from Columbia University

36 Business School. He was a primary developer of Betaflex, a statistical software package for

37 forecasting demand used by the Bell Operating Companies in the late 1980s and early 1990s. He

38 currently teaches econometrics and finance.

39 Q. On whose behalf are you testifYing?

40 A. I am testifying on behalf of ICSE-PR, the Institute for Competitiveness and Sustainable

41 Economy.

42 Q. Please summarize your testimony.

43 A. My testimony addresses the rate-making process from the perspective of someone who has a

44 good deal of experience evaluating key components of a successful rate filing process. This

45 experience includes organizing a rate-setting process, documenting it, responding to intervenors,

46 responding to Federal Communications Commission orders to make available underlying data,

21Page

47 methodology, and systems used to set rates. I have also had to develop systems for evaluating the

48 long-term effects of proposed regulatory refonn.

49 From my experience, I have learned that a successful rate-setting process relies heavily on

50 accurate historical data, accurate cost allocations to services, sound forecasting methodology,

51 which includes adjustments for known regulatory or market changes that can be documented and

52 assessed accurately. Systems need to be well documented and must pass audits by industry experts.

53 Documentation should include workpapers that explain each step in the filing process and should

54 be linked so that a reviewer can assess the consistency of the projections. A set of workpapers

55 should display differences between base period and test year demand, cost, and investment values.

56 It is also useful to have a sunrmary of historical and realized forecasts to measure forecast

57 performance.

58 My testimony begins with characterizing the proposed rate plan proposed by PREP A. It is

59 part of a comprehensive plan to restore PREP A to financial health and improve PREP A's service

60 performance. The overall plan assumes an agreement with creditor-approved revenue collections,

61 reduced operating costs, investment in new generation technology, increases in customer rates,

62 and a streamlined rate-setting process.

63 The rate plan is part of an overall strategy with multiple legislated objectives defined in

64 PREC Docket No. CEPR-AP-2015-0001:

65 • Provide for the combination of energy supply and conservation resources

66 that satisfies in the short-, medium-, and long-term the present and future

67 needs of the energy system both of Puetto Rico and of their customers at

68 lowest cost possible.

3/Page

69

70

71

72

73

74

75

76

77

• Promote and increase the use of clean and renewable energy through a

minimutn Renewable Portfolio Standard (RPS) of 20% by no later than

2035.

• Replacement of PREP A's outdated and inefficient technologies, cunently

generating electricity by burning fuel oil burning, over to a base of liquefied

natural gas (LNG).

• Intensification of distributed energy and smart grid technologies

• The implementation of Wheeling for electricity transmission and

distribution allowing for third party nse

entitled Specific Questions, I will examine the rate setting process from the

79 perspective of an outside party who is trying to assess if it is in conformity with best practices for

80 filing proposed rates. I am going to suggest that testimony I reviewed does not meet the standards

78 In the section

81 of a successful filing. As a result, I believe there are large risks associated with the rates proposed,

82 the trajectory of rates, the rate structure, and the resulting financial health of PREP A.

83 Many of these risks are of the type that any group of competent analysts would face. In no

84 way do I suggest that the cutTent group of analysts are not competent. In fact, their testimony

85 suggests that PREP A has gathered together a very good team to develop a rate-setting strategy.

86 The basic problems that they cannot avoid are as follows:

87

88

89

90

• Poor quality data that serve as the basis for rate-setting.

• The effects of rate shocks may cause large reactions of energy demanders that are

difficult to estimate. The rate shocks may also push PREP A away from its long

term distributed energy resomces (DER) goals and may encourage negotiated rates

4jPage

91

92

93

94

95

with large demanders of electricity (as stated m the proposed economic

development and load retention riders).

• The unknown effect of failing to reach an agreement with the remaining debt

holders to accept the debt restructuring plan.

• Even if the agreement is finalized, the risk of PREP A's debt rating not being raised

96 to investment quality is probably higher than the classification model suggests.

97 Important to note on the last two points, that although the issue of the restructuring bonds is part of a

98 separate proceeding, its success or failure in the markets will have an impact on the overall rate structure.

99 Given these challenges, I still believe there are other risks related to the limited number of

100 scenarios presented, the limited scope of the sensitivity analyses conducted, the lack of clearly

101 designed workpapers, and poor documentation of historical forecast accuracy. I am not convinced

102 that the proposed rate-setting strategy is consistent with satisfYing the objectives of PREP A's

103 Integrated Resource Plan (IRP) as modified by the Energy Commission.

104 I will document my concerns by examining the testimony submitted by several key

105 witnesses with the intention to raise questions about their assumptions and decisions made.

106 I will end with suggestions for improving the rate-setting process and review.

107

108

109

110

111

112

113

• It would be prudent to develop tariff filing modules and backup documentation that

would allow the Commission and interested pmiies to do scenario and sensitivity

tests.

• It would be prudent to have an action plan, in regards to tariff structure, in case the

current restructuring agreement falls apmi. Aclmowledging that, as previously

mention, although debt restructuring is separate proceeding its un-fulfillment would

have considerable effect on the rate structure. Having said so, the Restmcturing

5I Page

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

Bonds has not been issued at this time. This "what if' exercise may yield itmovative

financing proposals that may not have been considered before. For example, if

PREP A's fi"anchises were sold to private interests, or if other entities me allowed

to supply power generation in the near-tetm, especially those supplying power from

wind, solar energy and distributed energy in general, including efficient and clean

fossil generation, PREP A's need to raise funds would be reduced. Opening the

network to generating companies would have the added advantages of stimulating

local job growth and related economic activity that may be considered as indirect

funding.

• It would be prudent to recognize that PREP A's historical mismanagement suggests

that it may have a corporate culture that is not likely to meet customer needs. This

may continue to be true despite putting in new safeguards to prevent poor decision

making, and it may affect the perceptions of bond rating agencies despite an

increase in the company's debt coverage ratio.

• It would be prudent to consider this an appropriate time to consider eliminating

129 cross-subsidies in rates and establish an explicit universal service fund to fund

130 revenue shortfalls. Reinforcing the case for an explicit fund is the lack of

131 econometric models available to measure the loss of demand by customer class that

132 would result from the lmge proposed rate increases.

133 As a result of these suggestions the Commission should consider whether the revised data provided

134 by PREP A by October 11 to supplement or update the Petition for Rate Review is sufficient to

135 wanant the rate increases proposed by PREP A. If not, a separate proceeding should be considered

136 by the Commission to give interested patiies enough time to examine the rate-setting process and

6[Page

137 to assess whether the projected costs and proposed rates are reasonable, including to what extent

138 they are still inconsistent with the IRP; The separate proceeding should be more comprehensive

139 than the cun·ent proceeding. Besides following standard filing practices that tie rates to underling

140 base-year and projected costs and demands, it should also include historical forecast performance

141 measures, scenario and sensitivity analyses, and a history of data revisions.

142 SPECIFIC QUESTIONS:

143 Q. What is the focus of your testimony?

144 A. On a strategic overview of the current plan to reform PREP A's rate-setting process.

145 Q. What experience do you have in rate-setting?

146 A. I have been responsible for more than thirty rate-of-return (cost of service) filings in the

147 telecommunications industry and many mid-course conection filings. Conceptually, the rules and

148 filing documentation requirements for a successful telecommunications filing cany over easily

149 into the electric utility industry.

150 As a long-time participant in CRRl and now its director, I have been exposed to the work

151 of experts in the electric utility industry that focus on regulatory reform. I have been impressed

152 that many companies in the telecom and electric utility, and USPS face the challenge of declining

153 demand, large legacy rate-bases, and deteriorating long-term prospects.

154 It may be useful from the Commission's perspective to have a fi·esh view of the Rate Case currently

155 under consideration.

156 Q. How would you characterize the current rate-setting plan to revitalize PREP A?

157 A. I believe the plan has serious shortcomings that could lead to large unanticipated rate increases

158 without restoring PREP A to financial health. The rate plan is not consistent with opening the

159 network to new sources of renewable energy and to distributed generation in general. It also

7JPage

160 disadvantages industrial users with the consequences that they may try to bypass PREP A or

161 become less competitive in the marketplaces for their goods and services.

162 Q. What paris of the rate-setting process will you addTess?

163 A. I will examine (1) the data underlying the development of rates, (2) the cost of debt

164 methodology, (3) the modified cash flow method used to calculate revenue requirements, and ( 4)

165 the overall rate increase and rate design. In this last section, I will also discuss briefly the universal

166 service strategy based on cross-subsidies.

167 1. Data of unknown quality are being used to develop proposed rates

168 Q. How would you assess the quality of data being used to develop proposed rates?

169 A. I believe there are significant problems with the quality of data drawn from audited fmancials,

170 data needed from special engineering and other service cost studies, and from cost allocation

171 systems used to assign costs to service classes.

172 Q. What do you find deficient in the audited company financials?

173 A. Under normal circumstances, audited data provide reasonable assurance about the quality of a

174 company's data. See: http://accounting-simplified.com/audit/introduction/limitations-of-

175 audit.html. In PREP A's case,) do not have this assurance. I suggest that the Commission do a

176 comprehensive audit of all of PREP A's data systems that are required for setting rates, which

177 would include systems that generate standard financial statements that allocate costs to service

178 classes, that identify infrastructure replacement and upgrade needs, and that produce demand and

179 cost forecasts. This list is not meant to be exhaustive.

180 My testimony hinges on two statements made by PREPA witnesses, Lisa Donahue, Ralph

181 Zarumba, and Eugene Granovsky that strongly suggest PREP A may not have the billing systems,

8/Page

182 engineering systems, and other systems needed to identify costs and demands by service class,

183 which will handicap the rate-making process:

184 I begin with a statement made by Lisa Donahue followed by a two question raised by her

185 testimony. In (PREP A Ex. 2.0, p. 22 of 30) Donahue states

186 PREPA's [has had] operational challenges in recent decades, but with the additional

187 understanding that the work to implement the recovery plan has made progress (to varying

188 degrees) on these items

• A lack of institutionalized processes and procedures. 189

190 • Outdated systems and information teclmology ("IT"). Information systems that are

191 not integrated, resulting in duplicate data, poor data utilization, and poor data

192 quality. Underfunding of these systems, and frequent outages. Inadequate training

193 to effectively use and improve IT.

194 Q. Are you confident that the audited 2014 financials are up to the standards of a rate filing?

195 A. I am not sure. In PREP A's case, revenue data may be of poor quality because of inadequate

196 billing systems. To the extent this is true, any cost allocation based on revenue data may be

197 incmTect. For example, if bad debt expense is higher than booked levels, reported revenues will

198 be overstated.

199 Q. How would you assess the data used in special studies in light of Lisa Donahue's statement?

200 A. In my experience, even companies with dependable IT systems face a challenge extracting data

201 for special studies from financial and engineering records and combining them to develop special

202 studies. For example, forward-looking cost studies require management to evaluate the upgrades

203 to the electrical grid, maintenance and replacement of equipment, inventory, fuel purchases and

204 training employees to operate and maintain new equipment. Often the data gathering do not contain

9IPage

205 needed infmmation such as the cost of new equipment. As a result, special studies often require

206 incorporating new, extemal data such as the cost of new types of equipment to gauge forward

207 looking costs of services.

208 In my opinion, it is likely that special studies that rely on PREP A's IT systems will be deficient.

209 The next question is prompted by testimony given by Zarumba and Granovsky.

210 Q. Can you gauge the accuracy of cost allocations to service categories?

211 A. Again, the testimony by PREP A expetts gives little comfmt that the allocations are accurate.

212 Zarumba and Granovsky's testimony (PREP A Ex. 4.0, p. 3 of 59, line 56) states that

213 Cunently a very wide difference exists between allocated (embedded) costs as reflected in

214 the embedded cost of service study ("ECOSS") and the level of cost recovery of various

215 tariff classes (p. 3 of 59, lines 56 to 58)

216 Q. Does Navigant propose to adopt the revenue allocations produced in the ECOSS when

217 preparing the proposed tariffs?

218 A. No. We propose to mitigate the revenue allocations for the following reasons:

219 As previously discussed, many of the data inputs used in the ECOSS were estimated due

220 to data being dated or unavailable. Source: p. 23 of 59, lines 385 to 389.

221 Mitigated revenue or any cost allocation proxy based on revenue data may be biased if the revenues

222 are not generated by cost-based rates Moreover, as stated above, if the billing system is faulty it

223 will carry over to faulty cost allocations.

224 To the extent that coincident peaks are important for cost assignment, Zarumba and Granovsky's

225 testimony (PREP A Ex. 8.0, p. 16 of 20, line 292) give little comfort.

226 Q. Did PREP A have sufficient Coincident Peak ("CP") load data by Tariff Class to be used

227 as an allocation factor for assigning Production Demand and Transmission

10 I Page

228 Demand?

A. No. PREPA has limited coincident peale load data by Tariff Class. Navigant

investigated the extent that PREP A's metering systems (identified as MV90) collected

hourly load data. Navigant found that PREP A's load research information was insufficient

for ECOSS purposes for most of the Residential and Commercial Tariff classes.

229 Q. Are you comfmtable with the adjustments for known and measurable changes in financial data?

230 A. No. Although, adjustments for known and measurable changes in the financial data is a standard

231 procedure in rate cases, the issue here is the reliability of the adjustments. Parnpush et al (Ex. 5.0,

232 p. 20 of75, line 429) state that individual line items for 2014 financial data were adjusted based

233 on PREP A's restructuring consultant.

234 How did you arrive at your pro forma (!mown and measureable) adjustments?

235 We adjusted individual line items for what we conclude are known and measurable

236 changes. To arrive at these figures, we worked with PREP A's restructuring consultant,

237 AlixPartners, to dete1mine the estimated change in PREP A's costs of providing electric

238 service. We relied on the Business Plan and perfmmance improvement estimates

239 developed by AlixPmtners and PREP A to adjust non-fuel operations and maintenance

240 ("O&M") expense (Parnpush et al, PREP A 5.0, p. 20 of75, lines 429 to 435).

241 It is hard for me to judge the reliability of the adjustments because the testimony does not include

242 a track record for the restructuring consultant's forecast accuracy. Exercising due diligence

243 requires that this type of information should be disclosed. The Commission should require it.

244 2. Cost of Capital estimates are unreliable

245 Q. What problems have you uncovered about the cost of capital estimate used to develop proposed

246 rates?

111Page

247 A. To set rates, one needs a cost of capital target. In this case, the tmget return is unclear. Pampush

248 et a! simply chose the authorized rate of 12% ? (Ex. 5.0, p. 36, line 773). However, this rate is

249 well below cunent market based rates discussed on p. 35, which could be as high as 26.6% . (p.

250 35, line 760).

251 A pmdent approach to estimating the cost of debt would be to use a range of likely interest rates

252 so that an interested paTty can assess how sensitive rate levels would be to changes in the cost of

253 debt.

254 Funds availability now and in the future as well as the future PREP A bond rating will affect

255 the trajectory of rates and could force PREP A to miss its Integrated Resource Plan goals as

256 modified by the Commission.

257 In Exhibit #2 pages 13 to 14, lines 276 to 298 Witness Lisa Donahue stated that the

258 Restructuring Support Agreement (RSA) is not completed yet. It requires holders of $2B of

259 outstanding debt to agree to its terms. So fm, they have not. Moreover, failure to obtain an

260 investment grade rating for the Restructuring Bonds will go against requirements in the RSA. This

261 too is an open question:

262 Those economic terms, however, are not the end of the story under the RSA. There are

263 many noneconomic contingencies that must be met under the RSA, most of which are not

264 a part of the instant rate review. Those contingencies include, aTUong others, consent by

265 holders of $2 billion of the $2.7 billion in bonds that are not currently parties to the RSA,

266 the appointment of a new independent board for PREP A, approval of the calculation

267 methodology and true-up mechanism for the Transition Charge, obtaining an investment

268 grade rating for the new securitization bonds, successful completion of the validation

269 proceedings with respect to the Revitalization Act and the issuance of the new

12jPage

270 securitization bonds, and the continuing implementation of operational reforms leading to

271 cost savings.

272 Q. How would you evaluate the financial targets set to raise PREP A's bonds to investment grade?

273 A. I believe the projected Debt-Service-Coverage-Ratio (DSCR) produces an unwananted sense

274 of security that PREP A's bonds will be raised to investment grade. I believe the methodology used

275 to fix a tm·get suffers from a statistical problem called overfitting and more importantly fi·om a

276 missing variable problem.

277 Pmnpush et al (PREP A 5.0) projected the financial operating characteristics needed for

278 PREP A's bonds to become investment grade. They used a tree classification approach to determine

279 which financial ratios and their values m-e associated with investment grade bonds. They were

280 faced with the problem of using a dataset that had very few high-yielding bonds in it. They

281 analyzed 425 bond classification entries from 85 entities. (p. 51). Only 10 are high yield and only

282 25 more m·e BBB. (p. 63-75).In other words, they have very few observations to distinguish HY

283 bonds from other classifications.

284 They chose two stratification strategies to draw out bond-rating distinctions that

285 significantly reduce the number of usable observations. The stratification methods reported

286 reduced the sample size significantly (p. 63 of 75). Then they used sampling with replacement to

287 generate 1000 new random smnples. The results from the first simulation show that 9 of 10 HY

288 bonds had DSCR's less than 1.57 (p. 65 of75). The second method was dropped (p. 64 of75).

289 Re-stratifying the smnple to find a reliable model for classifYing bonds has the potential

290 shortcoming of overfitting the model. In other words, the final model chosen may appear more

291 precise than it is because the sh·atification method chosen may have been based on being the best

292 predictor of the test dataset. Moreover, the classification method, like other statistical methods,

BIPage

293 depends on a stable underlying process and identifying the correct causal variables. In this case, it

294 is likely that key variables are missing from the model. For example, PREP A's history of poor

295 management and the exe1tion of political influence to keep rates low may lead rating agencies to

296 require much stronger fmancial ratio targets than is typical of other entities.

297 It doesn't help PREP A's reputation when the Commission just rejected PREP A's IRP

298 (CEPR-AP-2015-0002) and issued a scathing assessment of PREP A's lack of compliance: The

299 Commission stated on page 4: "We are aware of no instance, among the many IRP proceedings in

300 the many mainland states, of such a large gap between regulatory expectation and utility response."

301 The Commission goes on to excoriate PREP A for a lack of transparency by not producing

302 workpapers, models, and modeling files, (p.7 and p. 47), monopoly mindset (p. 8) the lack of

303 scenario and sensitivity analyses (p. 11 and p. 47), poor track record of forecast accuracy (p. 15,

304 p. 42), among other complaints.

305 Pampush et al recognize the "missing variable" shortcoming.

306 [it] is unclear [that achieving this or any other benchmark will achieve investment grade

307 status] because the tme test of credit worthiness is the belief by lenders that they will

308 receive timely and complete repayments of all of the cash flows that are due to them. This

309 analysis cam10t capture these subjective beliefs, but can provide values relative to

310 quantitative indicators .... (p. 55 of75, lines 1166 to 1169)

311 3. The modified cash flow method for estimating revenue requirement will likely increase the

312 risk of uneconomic rate increases

313 Q. Should PREP A use the modified cash flow methodology for calculating revenue requirement?

314 A. A key feature of standard revenue requirement fmmulas is that the recovery of investments

315 occurs over the life of an asset. Rate payers in the current period do not pay for new plant and

14 I Page

316 equipment that will produce benefits for customers in future periods. The method for spreading

317 the cost of the equipment over the life of the new asset is to add depreciation expense to other costs

318 when calculating revenue requirement. For example, if a new asset costs $1000 and has a ten-year

319 life, and it depreciates in equal increments over its life, then revenue requirement is increased by

320 $100.

321 Parnpush et a! describe two standard methods that adhere to this recovery objective: the

322 accrual and cash basis calculations of revenue requirement. Parnpush et a! then recommend an

323 alternative approach called the modified cash flow method. This approach recognizes the entire

324 cost of the equipment in the year purchased. In my example. The $1000 expenditure on new

325 equipment would be added to other expenses to calculate revenue requirement in the current year.

326 Revenue requirement is typically on an accrual basis for investor-owned utilities as

327 Parnpush et a! acknowledge (P. 13 of 75). To display the differences in the three approaches,

328 Parnpush et a! display the formulas associated with each. The Accrual Basis approach has three

329 components: Operating Expenses, Rate Base, and Retum on Rate Base. The basic formula is

330 Revenue Requirement= Operating Expenses + Depreciation +

331 Rate Base X Rate of Return

332 Source: Parnpush eta!, (PREP A ex 5.0 p. 15 of75)

333 An alternative used by municipally-owned utilities is on a cash basis (p. 13 of 75)

334 The formula for the Cash Basis approach is

335 Revenue Requirement = Operating Expenses + Depreciation + Legacy Debt Service

336 +Additional Coverage to Meet Minimum DSCR + Gross Revenue Requirement for

337 Securitization

338 Source: Pampush eta!. (PREP A Ex. 5.0 p. 16 of75)

15IPage

339 The first two f01mulas recover the cost of investment in plant and equipment over their useful lives

340 through depreciation.

341 The modified cash flow increases revenue requirement in the current period by the full

342 amount of the capital expenditure. As a result, current rates must increase to recover the cash

343 outlays for new plant and equipment. The proposed formula is

344 Revenue Requirements= Operating Expenses (ex.Depr Expense) +Capital Expenditures +

345 Legacy Debt Service +Additional Coverage Requiredto Meet Minimum DSCR +Gross

346 Revenue Requirement for Securitization

347 Source: Pampush eta!, (PREP A Ex. 5.0 p. 18 of75)

348 As Pampush et a! admit:

349 In this approach, all costs are funded from cunent-period revenues, including funding for

350 necessary capital spending and the servicing of existing debt (Pampush et a!, PREP A Ex.

351 5.0 p. 18 of75).

352 The justification for recommending this approach is that PREP A does not have sufficient access

353 to capital markets to fund capital expenditures. [They] add capex directly to revenue requirement

354 rather than relying on a proxy such as depreciation or return on rate base. PREP A has significant

355 cash funding requirements for infrastructure improvements that cmmot be financed at this time,

356 but may be in the future. Adding capex costs directly will allow for the rate setting process to

357 consider only projects that PREP A needs to cash fund (Pampush PREP A Ex. 5.0 p. 18 of75).

358 Although the revenue requirement tables on p. 38 of75 show the tlu·ee approaches produce

359 very similm· revenue requirement estimates for FY 2014, this may not be true when PREP A makes

360 large investments in new plant and equipment. I suspect the modified cash flow method will

361 produce the highest revenue requirement. Pampush et a! note that "the proposed capital

16 I Page

362 expenditure program for the electrical system, which includes normal replacement of aging and

363 deteriorating plant, amounts to approximately $1.4 billion over the next three years (p. 40 of75).

364 Q. Are the long-te1m cost and investment projections produced by the witnesses for PREP A

365 reliable?

366 A. I am convinced are not reliable, especially after reviewing Exhibit 14.0, supplemental direct

367 testimony of Francis Pampush et al, that indicates it is not necessary or appropriate to change the

368 revenue requirement calculations previously filed Ex. 5.0 and related Schedules. Projections are

369 not reliable because they start with unreliable base period estimates. As a visionary exercise, it

370 lacks the scenario and sensitivity depth that would allow the Commission and other interested

371 parties to determine a least-cost path of future revenue requirements. Specifically, it ignores future

372 energy source alternatives and opening PREP A's network to competition, both of which would

373 reduce PREP A's funding requirements and instill a competitive spirit into an industry governed

374 by a monopoly mindset. Again, the exercise relies on internally generated forecasts that are not

375 documented.

376 Pampush et al begin with two stated objectives: for producing long-term revenue

377 requirement forecasts: to gauge the long-term trajectory of rates and revenues. The long-term

378 forecasts also help assess various restructuring options proposed by creditors and their advisors.

379 (Pampush PREP A Ex. 5.0 p. 39 of75)

380 The forecasts address four factors:

381 • Capital expenditures to suppmt a fuel change from oil to natural gas;

382 • Employee staffing reductions and operational improvements;

383 • Implementation of energy efficiency programs; and

384 • Potential changes in debt service requirements.

17 I Page

385 They consider only two scenarios: with restructuring and without restructuring (p. 42 of75). They

386 do not believe the no-restmcturing scenario is possible but cannot assign a probability to its

387 occunence.

388 We do not have a basis to assign a probability to such an outcome, but it certainly could

389 occur. (p. 43 of75)

390 Their long-term forecasts hinge on collected forecasts that are not documented.

391 [They are] based on infmmation from PREP A's financial depmiment, the Business Plan

392 prepared by AlixPminers and PREPA, and various financial obligation restructuring

393 scenmios provided by Millstein & Co, we developed a fifteen-year financial statement

394 forecast that includes the Income Statement, the Balance Sheet, Cash Flows a11d the Debt

395 Schedules. (p. 39 of75)

396 I have been involved in dockets where long-term forecasts are required. I will admit that "out-

397 yem·" forecasts m·e umeliable. Suppose a 15-year forecast was made for PREP A back in 2000, or

398 better yet, pick any historical yem, how accurate would it be? The problem here begins not with

399 the out-years; instead it is in the base yem.

400 In this case, the base-period assumptions of PaiDpush et a! me already out-of-date. The

401 Commission recently rejected the IRl' submitted by PREP A, which, I believe, serves as the starting

402 point for the fifteen-yem· forecast. Instead, the Commission issued a modified plan summarized

403 below.

404 • PREP A shall pursue permitting, engineering, and planning for AOGP, subject to a $15

405 million spending cap.

406 o New generation units:

407 o New combined cycle to replace the Aguirre steaiD units.

18 I Page

408

409

o Re-powering Aguirre combined cycle.

o New combined cycle units at Palo Seco.

410 • Retirements of oil-fired steam units

411 • Limited use of units for San Juan 9 and 10 units

412 • Improvements in transmission and distribution

413 • The implementation of new cost-efficient renewable energy projects in reference to

414 compliance with the RPS cunently required by law.

415 • Development "Energy Efficiency" Demand Response "and energy storage projects.

416 • Compliance with MATS

417 Source: Case No. CEPR-AP-2015-0002, Final Resolution and Order on the First Integrated

418 Resource Plan of the Puerto Rico Electric Power Authority

419 Moreover, any long-term projection should include other scenarios besides assuming the only

420 alternative fuel source is natural gas. It is inconsistent with the IRP's goals for renewable energy

421 and DER and projected investments. For example, one key IRP objective is to promote efficient

422 renewable energy projects.

423 One would expect that solar and wind generation projects would require new transmission

424 connections to the new energy sources. However, the table on p. 13 of21 ofZarumba's testimony

425 (PREP A Ex. 9.0) that displays ten-year totals for transmission investment shows zero investment

426 for interconnection of generation. This is a puzzling projection even in the face of declining load.

427 Another key objective is the "intensification of distributed energy and smrui grid technologies." I

428 am not able to connect the ten-year distribution investments displayed on p. 17 of 21 to this

429 objective. (Zarumba, PREP A Ex. 9.0). The Commission should consider these observations related

19 I Page

430 to Exhibit 4.0 when it reviews the adequacy of the revised data submitted by PREP A on October

431 11 to determine if the information is consistent with the revised IRP.

432 4. The proposed rate changes and rate design are based on faulty methodologies

433 Q. How would you evaluate the proposed rate changes and rate design?

434 A. Typically in a filing where the proposed overall rate increase is large, the filing entity projects

435 demand response to the rate increase. This allows fine tuning of projected revenues and hence the

436 proposed rates. This appear not to have been done. As a result, projected demand is likely

437 overstated. In the near-term bypass may result from the large proposed rate increases, but it has

438 not been factored into the revenue projections.

439 The rate-design also has significant shortcomings. It seems to focus more on meeting

440 financial targets than on efficient long-term pricing. For example, peak-load pricing is rejected.

441 However, it could be a way of incentivizing investments in solar and wind power by producing a

442 high payback during these periods. Moreover, PREP A witnesses do not explain the derivation on

443 non-volumetric fixed per month charges. This is likely a difficult exercise because transmission

444 and feeder plant are jointly used by customers. The method of recovering subsidy funds will

445 continue a practice of cross-subsidization that should be eliminated fi·om the rate-design. Instead,

446 funding for low-income residential customer programs should come from general tax revenues, if

447 possible.

448 In this type of turbulent environment, where realized costs are uncertain because of flawed

449 IT systems and large rate increases are proposed, the formula-based rate-change process may be

450 counter-productive. It may produce rate spikes that are not tied to the cost of providing service to

451 customer classes.

20 I Page

452 Q. Do you have any evidence that demand curtailment will result from the large proposed rate

453 increase?

454 A. The overall rate changes are large and not targeted towards customer classes in a cost-causative

455 manner. In this filing, rates are projected to increase, on average, by 26.5%.

456 PREP A's witnesses anticipated losses from price-sensitive customers. To mitigate rate shock, the

457 price increases are relatively uniform across service categories with the exception of public

458 lighting. See Zarurnba and Granovsky (PREP A Ex. 4.0, chart on p. 26 of 59). The rate increase

459 for industr·ial customers is much higher than warranted using allocated costs. As a backstop to

460 prevent customer bypass of PREP A, Zarumba and Granovsky recommend the "flexibility to

461 selectively discount tariffs is [sic] a verified risk can be demonstr·ated that load will be lost." (p.

462 36 of 59).

463 Major rate-change decisions are simply stated as a management decision. No rationale is given

464 based on any objective criterion, such as examining price elasticities. For example, Zarumba and

465 Granovsky state (PREP A 4.0, pp. 27-28)

466 Q. What tln·eshold was chosen to determine the maximum rate increase by customer

467 class?

468 A. It was determined that residential customers should not experience an increase that was

469 more than 5 percent in excess of that provided to the non-lighting customer classes. The 5

470 percent tln·eshold was chosen based upon judgment and reflects the opinions of PREP A's

471 management, the experience of the Navigant team, and socio-economic factors on the island.

472 Q. How do the targeted discounts reflect of the Fmmula Based Mechanism?

211Page

473 A. Selective relief to specific customers highlights a key flaw in the Formula Rate Mechanism

474 (FRM). In general, the assumption is that the FRM will pick up the revenue shmifall by raising

475 rates to all customers. This may create a spiraling loss of demand by customer class if it is the only

476 recovery mechanism available. It may make sense to have an alternate process that that allows for

477 a comprehensive reassessment of rates in the first year of its use and not wait for an overhaul at

478 the end of three years as proposed.

479 Q. What is your reaction to eliminating peak-load pricing?

480 A. Zarumba and Granovsky (PREPA 4.0) recommend that time of use pncmg should be

481 discontinued (p. 50 of 59). One justification is the increase in rooftop photovoltaic ("PV")

482 generation. This justification misses two key benefits of PV generation: it reduces the peak load

483 during high-usage periods and reduces the need for future generation capacity. Both effects should

484 reduce KV A charges now and in the future. A better approach is to continue offering the discount

485 and evaluate peale shifts periodically.

486 Q. How would you evaluate the proposed non-volumetric fixed charges?

487 A. It is not clear to me how the fixed charges were developed. The proposed tariff for GSS will

488 raise the fixed charge from $5.00 to $10.00 per month (PREP A 4.0, p. 47 of 59). The GSP fixed

489 charge will remain at $200 per month and new unbundled generation, transmission, and

490 distribution charges (p. 48 of 50). The witnesses do not describe how these non-volumetric charges

491 were detemrined. It is important to understand how the costs of jointly used facilities are distributed

492 among customer classes.

493 Q. What is your reaction to a volumetric subsidy charge?

494 A. It is not cost causative to assess a $.01020 kWh subsidy charge on GSP customers, for example

495 (PREP A 4.0, p. 48 of 59). This may be an appropriate time to consider a universal service fund

22 I P age

496 that is funded from general tax revenues in a similar fashion to funds raised for programs aimed at

497 aiding low income families.

498 Q. How would you evaluate having non-residential customers pay the transition charge estimated

499 to be $.03055?

500 A. Again, although this transition chaxge was appxoved, perpetuating cross-subsidies is not a good

501 long-term strategy. If a universal service fund were put into place, this burden of this obligation

502 could be more evenly spread.

503 Q. Overall, what is your strategic assessment of the rate-setting plan?

504 A. I believe it is a decent start for developing a comprehensive plan, but as it is proposed, it is

505 deeply flawed. The entire rate-setting process must be more transparent, calculations must be

506 reproducible by third parties, and the rates must be consistent with IRP goals. The Commission

507 should conduct an IT systems audit with the goal of developing modules that axe well documented

508 and easily usable to do scenario and sensitivity tests. The Commission should also consider a new

509 universal service fund to eliminate cross subsidies.

510 Q. Why is the current plan not in the long-term interests of Puerto Rico?

511 A. The rate-setting plan mainly focuses on preserving PREP A. The Commission should consider

512 adopting more stringent requirements for opening PREPA's Transmission and Distribution

513 network to competition. This approach will place less burden on PREP A for raising funds.

514 Competition may also reform PREP A's culture in ways that new internal policies and procedures

515 carmot.

516 I have great concem that customers are bearing more than their share of the financial

517 burden of decades of mismanagement. Much of the burden on customers is a direct result of

23 I Page

518 accumulating a high level of debt and this burden should be shared more equitably with

519 bondholders.

520 Q. What action plan would you recommend?

521 A. A comprehensive reassessment of the rate-setting process. The overarching goal is to develop

522 a rate-setting plan that is consistent with IRP objectives. A basic strategy for meeting IRP

523 objectives is to gradually replace PREP A as a vertically integrated monopoly with an open network

524 energy market. Towards that end, the Commission should consider developing a market for

525 generation and having a third party evaluate transmission needs. Strategic thinking is needed to

526 develop a plan for opening the distribution and transmission network to third-party generation,

527 including "behind the meter" and additional utility scale generation. It will also be necessary to

528 consider what entity or entities working cooperatively will have carrier-of-last resort obligations,

529 and how to pay for them. The long-term pricing strategy should accommodate these changes.

530 Within this context, the Commission should evaluate what is possible to achieve under the

531 current statutes and to consider amendments that would allow the Commission to achieve the

532 lowest cost stmcture for all customer classes. While global think is imperative, detailed-thinking

533 is also necessary for success. Rate-setting systems must be in place that are reliable, transparent,

534 and flexible. Billing systems, for example, should accommodate retail and wholesale rate structural

535 changes.

536 Q. Does this concludes your testimony?

537 A. Yes.

24 I Page

ATTESTATION

• J iltw'C ~ t:PS ;; Affiant,V · , being first duly sworn, states the following:

The prepared Testimony I am sponsoring constitute the direct testimony of Affiant in the

above-styled case. Affiant states that I would give the answers set forth in the Testimony

if asked the questions propounded therein at the time of the filing. Affiant further states

that, to the best of my knowledge, this statements made are true and correct.

t/tet-12/!-G-Iv~ {!

Affidavit No. ----

Acknowledged and subscribed before me by ~<i-f.OE.~ofthe personal

circumstances above mentioned, in his capacity as Consultant for the ICSE-PR., who is

personally known to me or whom I have identified by means of his driver's license

{{ tJ/02- 11.11'?'2- 1<:1 ~ numBer IJ2. lftfb , in /Jrv&ol?rflt iJ :J , thisLL_ day of

tJebtA .2016.

1

FERNANDO E. AGRAIT TESTIMONY ON

1 1. Question ("Q"): Please state your name and business address.

2 Answer ("A"): Fernando E. Agrait, Office 414, Centro de Seguros, 701 Ponce de Leon

3 Avenue, San Juan, Puerto Rico 00907.

4

5

2. Q:

A:

Please describe your current employment.

Since 2000, I have worked full time as a private attorney specializing in

6 commercial, banking, corporate, administrative law, government relation , and labor laws.

7 Since 1997, I have been Secretary of the Board of Directors of Centro Medico Del Turabo,

8 Inc. (now Grupo HIMA•San Pablo). Since 2000, I have served on the Boards of Directors

9 of Aireko Construction, as well as Antilles Cement Corp. and related corporations, among

10 others.

11

12

3. Q:

A:

Please describe relevant past employment and experience.

I have served in many academic positions at the University of Puerto Rico,

13 including President and Professor of Law. In other areas of public service, I have served

14 as Assistant Secretary and Undersecretary of the Puerto Rico Department of Justice,

15 Special Advisor on Legislative and Government Affairs in the Office of the Governor, Vice

16 President of the Puerto Rico Legislation and Jurisprudence Academy, Member of the

17 International Program Advisory Board of the National Science Foundation, Member of the

18 Advisory Committee on Civil Procedure of the Puerto Rico Supreme Court, and Member

19 of the National Advisory Board Sea Grant Program of the U.S. Department of Commerce.

20 I have also served as a Member of the Council on Puerto Rico-United States Affairs, a

21 Member of the Industrial Advisory Council, and Chair of the Governor's Adjunct Council

22 on Science and Technology.

1

23

24

4. Q:

A:

Please describe your educational background.

I received a Bachelor Degree in Business Administration with a major in

25 accounting and a J.D. from the University of Puerto Rico, as well as an LL.M from Harvard

26 University. In addition, I received Honorary Doctorates from the City University of New

27 York and the University of South Carolina.

28 5. Q: On whose behalf do you appear in this proceeding?

29 A: Institute de Competitividad y Sostenibilidad Econ6mica de Puerto Rico

30 (ICSE-PR).

31 6. Q: What is the purpose of your testimony?

32 A: My testimony describes the applicable law to the rate proceedings, identify the

33 value and interests mandated to be protected by Law 57 of 2015, its interrelation with the

34 rate and IRP proceedings and conclude that PREPA's rate submittal ignores and does

35 not comply with the Law 57 requirements and is intrinsically inconsistent with such

36 mandates.

37 My testimony also concludes that, based on PREP A's own admissions, information

38 provided, and recognition of information and studies still lacking, has not carried the legal

39 burden to obtain approval for the rate submittal.

40 7. Q: According to your position which is the applicable law and what are the

41 mandates for the rates proceedings?

42 A: The applicable law is Law 57 of 2015. Law 57, contrary to Law 4 of 2016,

43 which was the main applicable law for the transition charges and securitization process,

44 does not in any manner limit the discretion and power of the Energy Commission (PREC).

2

45 On the contrary Law 57 states very clearly mandates of law that refer to how PREP A

46 should operate and the specific values and interest to be protected.

47 With great clarity Act 57 identifies how "costly" electricity service is in Puerto Rico,

48 which "impedes economic development" and hinders efforts to stimulate the economy.

49 The act's preamble describes Puerto Rico as "hostage to an inefficient energy system."

so To address this problem (in addition to reducing air contamination), Act 57 aims to

51 transform PREPA, move the Commonwealth to save energy consumption; promote net

52 metering and renewable energy; establish "regulation" to promote the use of " highly

53 efficient fossil generation", based on an integrated resource plan with a 20-year horizon.

54 Furthermore, the Puerto Rico Legislature and the Governor, committed to Act 57's goals

ss of "Puerto Rico Energy Transformation and RELIEF" with the broadest citizen

56 participation.

57

58

59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76

Power:

Section 1.2 of Act 57 provides the following Public Policy Statement on Electric

"(a) The cost of the electric power generated, transmitted, and distributed in Puerto Rico shall be affordable, just, and nondiscriminatory for all consumers; (b) The availability of energy supply shall be guaranteed to the People; (c) The implementation of the public policy on energy shall be an ongoing planning, consultation, execution, evaluation, and improvement process in all energy-related matters; (d) The implementation of strategies geared toward achieving efficiency in the generation, transmission, and distribution of electric power shall be sought in order to guarantee the availability and supply thereof at an affordable, just, and reasonable cost; (e) The safety and reliability of the electricity infrastructure shall be guaranteed by integrating clean and efficient energy and using modern technological tools that promote economic and efficient operations; (f) The electrical infrastructure shall be maintained in optimum conditions as to ensure the reliability and safety of the electric power service;

3

77 78 79 80 81 82 83 84 85 86 87

(g) The Island shall become a jurisdiction with diversified energy sources and high efficiency electric power generation. To achieve this, it shall be necessary to reduce our dependence on energy sources derived from fossil fuels, such as oil, and to develop short-, medium-, and long-term plans that allow us to establish a well-balanced and optimum energy portfolio for the electrical system of the Commonwealth of Puerto Rico; ... "

88 " .. (m) Prices shall be based on the actual cost of the service 89 provided, efficiency standards, or any other parameters 90 recognized by government and non-governmental 91 organizations specialized in electric power service; ... " 92 93 In the same manner, the following language of Section 2.6 summarizes the

94 mandate of Law 57:

95 "The Authority shall rise to energy and environmental 96 challenges by using scientific and technological advances 97 available; incorporate the best practices in the electric power 98 industries of other jurisdictions; make the connection of 99 renewable energy producers to the electric power grid

100 feasible; carry out any process needed to make the electric 101 power generated in Puerto Rico, whether by PREPA, co-102 generators, or independent power producers, highly efficient 103 and clean for a better environment and public health." 104

105 8. A: Why do you state that PREP A's rate proposal does not comply with Law

106 57 mandates.

107 Before answering this question I would like to state our respect for PREPA

108 Executive Director Dr. Javier Quintana and the team of analysts involved on this

109 proceeding, since in no way the ICSE-PR or do I suggests incompetency as part of this

110 filing, however, the issues before this Commission are very complex and need careful

111 and comprehensive analysis. In this regard I would also like to express my respect to the

4

112 PREC which has demonstrated a high degree of seriousness in handling these difficult

113 issues.

114 Liza Donahue has stated:

115 "As CRO, my fundamental role is to lead the development and 116 implementation of a holistic recovery plan that allows PREP A to serve the 117 people of Puerto Rico more efficiently, reliably, and at a reasonable cost. 118 The recovery plan, at the highest level, involves two tasks: (1) addressing 119 the fundamental operational problems that have hindered PREPA for 120 decades and (2) financial restructuring in light of PREP A's debt and liquidity 121 crises". 122

123 There is, on the testimony no mention of legislated mandates, except at p. 2, but

124 without any content.

125 She continues:

126 "As CRO, I have led PREPA's negotiations with creditors 127 holding, in the aggregate about 70% of PREP A's debt. The 128 negotiations have culminated in a lengthy and detailed Restructuring 129 Support Agreement (as amended or restated from time to time, the 130 'RSA"), which provides for the restructuring of PREPA's financial debt 131 subject to the satisfaction of certain conditions and milestones. The 132 foundation of the RSA is PREP A's recovery plan, which requires 133 equitable burden sharing by all stakeholders, inc! uding PREP A's 134 management and employees through labor and operational 135 savings, governments and municipalities through timely payment of 136 energy bills and reform of the CIL T system, and PREP A'; creditors 137 though creating a sustainable capital structure. Throughout this 138 process, we have been very clear that PREP A's recovery plan cannot 139 be built solely on the backs of ratepayers. It must be an 140 equitable process with shared burdens and shared benefits." 141 142 It is Ms. Donahue and her vision of PREPA bending over backward to avoid

143 recognizing and complying with direct specific legislative mandates.

144 Donahue adds:

145 "As result, absent a successful financial restructuring and addressing 146 the liquidity issues, no version of PREPA, no matter how much its 147 operations are improved, can succeed."

5

148

149 The problem with these statements is that it look to the PREPA situation 150 mostly from the financial perspective, but does not look at it at all from the legislative 151 mandated perspective. It is a two way street.

152 Donahue continues: 153 154 "those economic terms however, are not the end of the story under the 155 RSA. There are many noneconomic contingencies that must be met 156 under the RSA most of which are not a part of the instant rate 157 review. Those contingencies include, among others, consent by 158 holders of $2 billion of the $2.7 billion in bonds that are not currently 159 parties to the RSA, the appointment of a new independent board for 160 PREPA, approval of the calculation methodology and true-up 161 mechanism for the Transition Charge, obtaining an investment grade 162 rating for the new securitization bonds, successful completion of the 163 validation proceedings with respect to the Revitalization Act and the 164 issuance of the new securitization bonds, and the continuing 165 implementation of operational reforms leading to cost savings. All of those 166 contingencies are works in progress. Failure to meet any of the milestone 167 and other conditions precedent could result in te1 mination of the RSA 168 and the loss of nearly $1.4 billion in liquidity savings and debt reductions 169 embodied by the RSA." 170 171 Ms. Donahue identifies the issues but fails to offer or present any option. There is

172 no mention at all of private direct investment.

173 Finally, Ms. Donahue states: 174

175 "Yes. For many years, PREPA has been run by successive Governing 176 Boards and senior management teams that have been subject to the 177 changing direction and policies of different administrations. Management 178 and other strategic decisions, including staffing and capital investment, too 179 often have been based on political considerations rather than best practices 180 or sound business judgment." 181 182 From 1992 to 2012, 20 years PREPA was under the same political party control

183 that is 60% of the time. The political control issues is a "red hering". The PREP A problem

184 is the incredible consistency of its erroneous and failed policies, mostly based on an

185 internal" monopoly know all" mentality, which the IRP order of this Commission identifies

186 and tackles.

6

187

188 10. Q: Please be more specific on the Law 57 values and interests protected.

189 Law 57 establishes very clear mandates on what are the values and interests

190 promoted and protected by the PREPA transformation. The law states the specific

191 mandates, and enumerates the very specific interests and values it wants to promote and

192 protect. All these interests and values are either mandated or are the result and

193 consequence of new structures (The Commission), procedures and information

194 requirements in the law.

195 More than a mere enumeration of aspirations, the law is very specific in its

196 mandates and such mandates must be part of the process of evaluating the new rates.

197 The Statement of Motives expresses:

198 a. There is a broad consensus on the need to evolve from our dependence 199 on fossil fuels and use to the maximum extent possible the Island's 200 energy resources, such as the sun and the wind, conservation, and 201 efficiency. 202 203 b. The high cost of energy limits our ability to stimulate the economv. 204 strengthen small- and medium-sized business, as well as to attract 205 private investors from abroad, develop commercial, industrial and 206 manufacturing activities, and improve the quality of life of all Puerto 207 Ricans. This prevents our Island from becoming a competitive and 208 attractive place in all aspects. We have been held as hostages of a 209 poorly efficient energy system that excessively depends on oil as fuel, 210 and that does provide the tools to promote our Island as a place of 211 opportunities in the global market. The current cost per kilowatt-hour of 212 approximately twenty-seven cents ($0.27) is extremely high when 213 compared to other jurisdictions that compete with Puerto Rico to attract 214 investors and severely affects the pockets of local consumers. 215 216 Nowadays, at an all times low oil cost, the average base cost per kilowatt-hour of

217 approximately eighteen cents per kilowatts-hour ($0.18/KWk) near double of the

218 national US average and considerable over other jurisdictions that compete with

7

219 Puerto Rico to attract investors and severely affects the pockets of local

220 consumers. This rate does not consider the temporary increase approved by the

221 CEPR in June, 2016 of 1.3 Cent/KWh, neither the transition charge of 3.1

222 Cent/KWh. If those amounts are considered, the rate would be 22.4 Cent/KWh. In

223 addition, as fuel price expectedly increases, electricity cost will also increase.

224 c. Therefore, it is imperative and compelling to enforce a thorough reform of 225 the energy sector that promotes the operation and administration of an 226 efficient system at just and reasonable costs, considering that we are an 227 isolated jurisdiction that needs to have a safe and stable electric power grid. 228 We need to adopt a regulatory and legal framework through the creation of 229 a robust independent entity that will ensure the transformation of the electric 230 power system of our Island for the benefit of present and future generations. 231 232 d. After more than seventy (70) years of its creation, and more than three 233 decades of having achieved the total electrification of the Island, PREPA 234 has become a monopoly that regulates itself; sets its own rates without 235 actual oversight; incurs operational, managerial, and administrative 236 deficiencies whose actual cost, at the end of the day, is borne directly by 237 customers; and whose governance lacks transparency and citizen 238 participation. All of the above contributes to Puerto Rico being among the 239 top U.S. jurisdictions with the highest energy cost. 240 241 e. As a summary the Statement of Motives states: 242 243 This Legislative Assembly reasserts its commitment to the People of Puerto 244 Rico through the creation and implementation of an Energy Reform 245 consisting of multiple initiatives that are all related to common goals such 246 as permanently reducing the cost of energy and provide the People of 247 Puerto Rico with a reliable, affordable, efficient, and transparent electric 248 power service. 249 250 f. P. 9- At that time, our Legislative Assembly stated that Act No. 114-2007 251 resulted, among other things, from the need to incentivize the generation of 252 electric power through renewable energy sources due to our excessive 253 dependence on fossil fuels to generate electricity and their well-known 254 polluting effect on the environment, as well as the high costs reflected on 255 electricity bills. However, in spite of the subsequent amendments to the Act, 256 the regulations adopted by PREP A with regard to systems with a nameplate 257 capacity in excess of 1 MW, far from supporting the development of 258 renewable energy alternatives, have had the practical effect of obstructing 259 the development thereof.

8

260 261 g. Section 1.2 of Act 57 states in part: 262 263 (1) The safety and reliability of the electricity infrastructure shall be 264 guaranteed by integrating clean and efficient energy and using modern 265 technological tools that promote economic and efficient operations; 266 267 (2) The electrical infrastructure shall be maintained in optimum conditions 268 as to ensure the reliability and safety of the electric power service; 269 270 (3) The Island shall become a jurisdiction with diversified energy sources 271 and high efficiency electric power generation. To achieve this, it shall be 272 necessary to reduce our dependence on energy sources derived from fossil 273 fuels, such as oil, and to develop short-, medium-, and long-term plans that 274 allow us to establish a well-balanced and optimum energy portfolio for the 275 electrical system of the Commonwealth of Puerto Rico; 276 277 (4) Prices shall be based on the actual cost of the service provided, 278 efficiency standards, or any other parameters recognized by government 279 and nongovernmental organizations specialized in electric power service; 280 281 (5) Every electric power contributions, subsidies, or direct or indirect 282 payments provided by PREPA shall be properly used in accordance with 283 the objectives for which they were granted. 284 285 (6) "Interconnection Charge"- Shall mean the fair and reasonable amount 286 of money that a person shall pay to PREP A for the right to connect his/her 287 facility to Puerto Rico's electric power grid. 288 289 (7) "Wheeling Rate" - Shall mean a just and reasonable amount of money 290 that PREP A may charge to a power producer for using its transmission and 291 distribution facilities for wheeling and for the right to interconnect the 292 electrical power generation facility of such power producer to the electric 293 power grid of Puerto Rico, in accordance with the provisions of Act No. 73-294 2008. 295 296 Article 2.9 (a)(viii) 297 (viii) Distributed Generation - PREPA shall identify the most effective and 298 economic manners to make the electric power infrastructure of Puerto Rico 299 more distributed and sustainable, and promote the use and strategic 300 integration of sustainable energy technologies and practices. To comply 301 with this duty, PREPA shall plan, build, and update distribution systems to 302 ensure the integration to the maximum extent possible of renewable 303 distributed generation. 304

9

305 These are just some of the values mandates to be protected by Law 57.

306 11. Q: Why do you state that PREPA's proposal does not protect or even

307 acknowledge these legislated mandates?

308 A: First there is no mention of those mandates, except in a very perfunctory

309 manner, without any content or context.

310 Second: The PREC made very strong, very specific determination concerning

311 PREPA's noncompliance in the IRP proceeding, which certainly impact the rate

312 proceedings.

313 Third: Even after PREC IRP order, which stated at page 4.

314 "We are aware of no instance, among the many IRP proceedings in the 315 many mainland states, of such a large gap between regulatory expectation 316 and utility response." 317

318 PREPA's reaction is:

319 A) As at expressed by Quintana at page 1 of 12, October 14, 2016 supplemental direct

320 testimony:

321 "As I explain, PREPA does not believe that the IRP Order will have a 322 material or quantifiable effect on its actual FY2017 revenue requirement, 323 especially given the additional assumptions concerning AOGP that the 324 Commission has asked us to make. Also, PREP A cannot determine at this 325 time what the ultimate effect of the IRP Order and the orders what will follow 326 over the next three years will have on PREPA's FY2017M FY2018 and 327 FY2019 business plans under those same assumptions. Based on what 328 we know now and on the assumptions identified in the September 29 Order, 329 we cannot be sure whether the IRP Order will increase or reduce PREP A's 330 costs." 331

332 And at page 6 of 12:

333 "PREP A does not believe that the IRP Order materially effects the revenue 334 requirement at issue in this rate case." 335

10

336 B) As expressed by Hemphill October 13, 2016 supplemental testimony at page 2 of 17:

337 "I do not intent to change anything fundamental about my proposal, but 338 rather to provide added detail (e.g., regarding dates of various filings, what 339 information would be filed, the the nature of certain proceedings, and when 340 they would occur) to help explain why this particular FRM is in the interests 341 of Puerto Rico and can, as I said, effectively and fairly respond to the 342 changes and uncertainties the Commission has identified." 343 344 C) Pampush, Porter and Stathos, October 13, 2016 supplemental testimony state at page

345 1 of 5:

346 "The purpose of this supplemental testimony is to 1) respond to the 347 Commission's Order dated September 27, 2016 ("September 27 Order") 348 requesting PREP A to provide revisions to the revenue requirement to reflect 349 certain considerations consistent with the Commission's Integrated 350 Resource Plan ("IRP") Order of September 23, 2016 ("IRP Order"); and 2) 351 to provide errata to previously filed testimony." 352

353 And answer at page 2 and page 4:

354 Q. Is it necessary or appropriate to change the revenue 355 requirement calculations previously filed as PREPA Ex. 5.00 or the 356 Schedules thereto in response to the IRP Order? 357 A. No. The Commission's September 27 Order calls on PREP A 358 to revise its testimony and exhibits as required by three considerations. 359 The first is that "the costs reflected in the revenue requirement shall be 360 consistent with the Modified IRP established in Part VII of the IRP Order." 361 The IRP Order does not materially change, or change in a known and 362 measurable way, PREPA's adjusted test year revenue requirement. 363 PREPA reports that the direct changes to PREPA's IRP-related 364 expenditures through the June 30, 2017 will be modest and primarily 365 relate to planning for different resource and system contingencies, 366 costs for a type of work that is likely to be equally required regardless of 367 what specific investments are approved in the IRP Order.

368

359 "Q. Is the answer any different with respect to the business plans 370 for FY2017, 2018 and 2019? 371 372 A. No. The logic applicable to the FY2017 revenue requirement 373 continues to apply.

11

374 Moreover, the years beyond FY2017 include significant 375 investment in AOGP and repowering oil units to use natural gas with a 376 resulting forecast reduction in fuel expense."

377 12. Q. What is your position, concerning the relation of PREC IRP Order and the

378 rate proceedings?

379 A. The IRP order is very specific. The Commission made very specific statements

380 concerning PREP A's noncompliance with the law and the Commissions' order and made

381 very specific requests.

382

383 The PREC stated in the order. Among other determinations.

384

385 1. This final Resolution and Order will affect the revenue requirement

386 PREP A proposed in Case No. CEPR-AP-2015-0001, In Re: Puerto Rico

387 Electric Power Authority Rate Review. In the proceeding, the Commission

388 will direct PREPA to make the necessary adjustments to its Petition,

389 testimony and exhibits.

390 2. Regarding the requirement to develop a range of viable

391 alternative, IRP did not comply. Contrary to an explicit Commission

392 rule, PREPA failed to use a capacity expansion model. This failure

393 precluded both a reasonable exploration of alternative resource

394 portfolios and a rigorous means of optimizing them. PREPA's

395 methodology and the range of alternatives it produced are

396 unsatisfactory.

397 Regarding the requirement to forecast load, the IRP did not

398 comply. PREPA failed to compare historical actual load with historical

399 forecasts, failed to substantiate adequately how it developed its

400 forecasts, and did not explore a reasonable set of future uncertainties,

12

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

1.

particularly a lower demand consistent with anticipated energy efficiency

programs and other factors.

Regarding the requirement that the !RP be consistent with

statutory objectives with respect to energy efficiency, renewable

energy, and the performance of fossil fuel generation, PREPA did

not comply. PREP A assumed incomplete compliance with the RPS

and failed to consider 100% compliance with the governmental

energy efficiency program.

For Puerto Rico, the goal, in short, is to replace old, costly plants

with lower-cost options: more efficient plants, renewable resources,

energy efficiency, demand response and distributed generation

technologies- some of which empower consumers to manage their own

costs, all of which reduce environmental damage as well as customers'

exposure to fuel price volatility. Properly designed and continuously

executed, an integrated resource planning process will carry out the

Legislature's intent to evolve the energy sector into one that relies less

on imported fossil fuels and more on Puerto Rico's own resources.

Finally, Intervenors raised several issues that are important to

consumers but are outside the statutory scope of integrated resource

planning. One of such issues is hedging strategies, which are relevant

to PREP A's near-term procurement plans but are not part of a long-term

planning process. For example, such strategies should be examined in

rate cases, where we determine how fuel costs (including the cost of

13

424 hedging strategies) should be accounted for in PREPA's rates. Other

425 issues raised by Intervenors were retail wheeling, competitive supply,

426 and generation ownership. These issues all relate to market structure,

427 i.e. the types of sellers and buyers of electricity services, the

428 mechanisms by which they offer and are paid for such services, and

429 their responsibilities as market participants. While a competitive

430 electricity market might offer value in Puerto Rico, an IRP proceeding is

431 not the appropriate forum to evaluate that option. An IRP proceeding

432 addresses the mix of power resources that is appropriate for Puerto

433 Rico. The types of market structures mostly to produce that mix cost-

434 effectively is a subject for separate proceedings.

435 2. Fossil fuel vs. renewables and demand-side

436 management: In its IRP filings, PREP A did not explicitly examine a

437 tradeoff between new or existing fossil generation on the one hand

438 and renewable energy or energy efficiency on the other. Rather, PREP A's

439 IRP filings incorporated renewable energy to the extent required to

440 satisfy its assumed RPS targets. Put another way, PRE PA viewed

441 renewable energy as a legal obligation to satisfy, rather than a

442 resource that could compete with fossil-fired generation. PREP A also

443 assumed that new renewable energy would involve a cost comparable

444 to PREP A's currently held renewable energy contracts. The Revised

445 IRP did not consider demand-side management- either energy

446 efficiency or demand response-for peak reductions.

14

447 Energy efficiency, demand response and energy storage shall be

448 treated as follows:

449 a. PREPA shall cooperate with one or more Commission-

450 appointed third-party administrators in the preparation of a potential

451 study (or studies if necessary) regarding energy efficiency and

452 demand response.

453 b. PREPA shall cooperate with one or more Commission-

454 appointed third-party administrators in designing and implementing,

455 cost-effective, state-of-the art energy efficiency and demand

456 response programs.

457

458 c. PREP A shall submit an updated analysis of the potential

459 for energy storage in the next IRP (unless the Commission order

460 such analysis to be produced at an earlier date), with updated costs

461 assumptions and an improved modeling approach and valuation

462 methodology.

463

464 What it is amazing is PREP A's reluctance to understand PREC's clear directives,

465 in addition to its no compliance with Law 57 mandates.

466 When PREP A was specifically asked by ICSE-PR on how the rate proposal and

467 the IRP interact with the law mandates, PREP A answered (ICSE-PR Request No. 24 and

468 25).

469 470 "PREPA objects to this Request as including factual assertions and legal

15

471 claims and arguments as given in the questions, unreasonable, vague, over 472 broad, unduly burdensome, calling for speculation, requiring of PREPA to 473 create significant analyses not currently in existence, and as improperly 474 calling for legal analyses. Without waiving its objections, PREP A states as 475 follows. PREPA's rate fillings (original and supplementary) must be taken 476 as a whole, and PREPA's rate filings are intended to comply with all 477 applicable legal standards. On July 15, 2016, after PREP A provided certain 478 supplementation, the Energy Commission issued an Order and Resolution 479 finding the filings complete for purposes of Reg. No. 8720. The Commission 480 ultimately will use the final Order in this rate review, making such findings, 481 conclusions, and rulings as the Commission concludes are warranted by 482 the record and the applicable law." 483

484 "PREPA objects to this Request as vague, unreasonable, over broad, 485 unduly burdensome, calling for speculation, requiring of PREP A to create a 486 significant analysis not currently in existence, and as improperly calling for 487 a legal analysis. Without waiving its objections, PREPA states (1) that it 488 filed testimony and other documents regarding how the proposed IRP 489 related to the rate filing; (2) the IRP final Order was issued on September 490 23, 2016; and PREPA, per Commission order, soon will provide 491 supplementation on this subject." 492

493 Answer to Request 30 and 31 are: 494 495 "PREPA objects to this Request as vague, unreasonable, over broad, 496 unduly burdensome, calling for speculation, and as improperly calling for a 497 legal analysis. Without waiving its objection, PREPA states as follows. 498 PREPA's rate filings (original and supplementary) must be taken as a 499 whole, and PREPA's rate filings are intended to comply with all applicable soo legal standards." 501

502

503 "PREPA objects to this Request as including legal characterizations as 504 given in the questions, unreasonable, vague, over broad, unduly sos burdensome, and as improperly calling for legal analyses. Without waiving 506 its objections, PREPA states as follows. PREPA's rate filings (original and 507 supplementary) must be taken as a whole, and PREPA's rate filings are 508 intended to comply with all applicable legal standards. On July 15, 2016, 509 after PREPA provided certain supplementation, the Energy Commission 510 issued on Order and Resolution finding the filings complete for purposes of 511 Reg. No. 8720. The Commission ultimately will use the final Order in this 512 rate review, making such findings, conclusions, and rulings as the 513 Commission concludes are warranted by the record and the applicable law.

16

514 The subject of renewables is discussed in the rate Petition and in multiple 515 testimonies and other filed documents. See, e.g. Petition, pp. 1-2, 'fl1f1, 3, 516 8, 21' 24, 73." 517

518 Incredibly in these answers PREP A refer to the rate and IRP fillings when at the

519 time the questions where asked and the answers given, the PREC had already

520 specifically rejected the filed IRP.

521 PREP A has a fixed mentality and has determined that going back to the market in

522 2020 is the most important if not the only important consideration. Within that "conceptual

523 lock", PREP A tries to fit all the available facts. As Professors Stephen J. Gould said: "it

524 is not factual lack, but conceptual lock".

525 13. Q. What issues or important considerations are not been taken into account.

526 A. First. There is no consideration of the elasticity of demand to evaluate the impact

527 of a significant rate raise.

528 Second: there has been no effort to study by PREP A the economic impact of such

529 increase, not only in inflation rate, but also in terms of the economic growth. ICSE has

530 submitted a separate testimony by Economist Ramon Cao which covers this important

531 issues and demonstrate the lack of economic analysis by PREP A.

532 Third: there has been no effort to quantify different impacts in different sectors.

533 While PREPA accepts that there is a limit to how much residential rates can increase,

534 there is no analysis of what increase different economic sector can tolerate.

535 (See Zurumba October 13,2016, P. 6 of 15)

536 Fourth: : The possibility of substituting PREPA investments, presently based on

537 obtaining the required revenues from an increase in rates, or in a future return to the

538 markets, with private direct investment is simply not even considered.

17

539 Fifth: The real actual possibility of the whole processes of rate and securitization

540 to unravel, due to unsustainable increases is not even mentioned.

541 Sixth: The Law 57, and IRP Order mandate to consider demand side, energy

542 efficiency and increased renewables solutions. PREPA does not even mention these

543 issues.

544 Seventh: there is no discussion at all that the rate increase, cost allocation per

545 costumer class, proposed riders and the overall tariffs new situation represent in regard

546 to promoting or serving as an incentive to simply abandoning the grid.

547 The unsustainable increase in price, together with the proposed reduction in the

548 net metering credit and increases in fixed costs, together with rapid and continuous

549 improvements, in renewable production and energy storage in batteries, are a recipe for

550 promoting abandoning the grid. PREP A does not even considerer this issue.

551 Eighth: The "Iasser like" focus on going to the market in 2020 does not consider

552 whether the decisions taken to reach such goal are simultaneously creating conditions

553 that will make more- not less- difficult to reach it. Especially when PREPA lacks any

554 coherent explanation on the impact of population reduction, nonexistent economic growth

555 and the 10 years of continuous reduction in sales. PREP A, as previously mentioned,

556 does not consider the economic impact of the rate increase which can further reduce

557 demand, lower sales and be at the end self-defeating.

558 Nineth: PREP A does not address the impact, importance and obligation to provide

559 for wheeling. Wheeling which under Law 73 of 2008 was limited to renewables, under

560 Law 57 is open to nonrenewable energy.

561 Tenth: The analysis of renewable, is considered only as direct price economic

18

562 issue and does not evaluate non direct economic benefits like cleaner environmental, less

563 dependence on oil, gas or any other external source, and more local control of energy

564 sources.

565 Eleventh: The elimination of TOU and the increase of fixed vs. variable costs has

566 non evaluated impacts on demand side control and on the viability of the renewable

567 energy industry in Puerto Rico. Such changes reduce incentives to reduce peak hour

568 load and are a disincentive to reduce load to PREP A by installing renewable alternatives

569 and energy efficient equipment and practices.

570 Finally: there is no mention, in any supplemental PREPA document, of the

571 possibility of rethinking its financial strategy and plans in light of the PROMESA

572 legislation.

573 For the reasons above stated, it is our opinion that this Honorable Commission,

574 should disapprove the proposed rate increase and perform a comprehensive

575 reassessment of PREP A's petition. In a complementary manner, the Commission should

576 refer to the Puerto Rico Oversight Board the findings under this proceeding, with the

577 purpose of a deeper view of PREP A's finances and further restructuring of its obligations,

578 including its legacy debt.

579 12. Q. Does this complete your testimony?

580 A. Yes, thank you.

581

582

583

584

19

585

586 Affidavit No. __ B_'1_0_

587 Sworn before me by Fernando

588 resident of Guaynabo, Puerto Rico to whom I known personally, today 25 of

589 October of 2016. In San Juan, Puerto Rico.

590

591

20